1 Genomic Diversity and Global Distribution of Saccharomyces Eubayanus, the Wild Ancestor Of

bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

1 Genomic diversity and global distribution of Saccharomyces eubayanus, the wild ancestor of

2 hybrid lager-brewing yeasts

4 Quinn K. Langdon1, David Peris1,2,3, Juan I. Eizaguirre4, Dana A. Opulente1,2, Kelly V. Buh1,

5 Kayla Sylvester1,2, Martin Jarzyna1,2, María E. Rodríguez5, Christian A. Lopes5, Diego

6 Libkind4,@, Chris Todd Hittinger1,2,@

8 1Laboratory of Genetics, J. F. Crow Institute for the Study of Evolution, Wisconsin Energy

9 Institute, Genome Center of Wisconsin, University of Wisconsin-Madison, Madison, WI 53706,

10 USA

11 2DOE Great Lakes Bioenergy Research Center, University of Wisconsin-Madison, Madison, WI

12 53706, USA

13 3Department of Food Biotechnology, Institute of Agrochemistry and Food Technology (IATA),

14 CSIC, Valencia, Spain

15 4Laboratorio de Microbiología Aplicada, Biotecnología y Bioinformática de Levaduras, Instituto

16 Andino Patagónico de Tecnologías Biológicas y Geoambientales (IPATEC), Consejo Nacional

17 de Investigaciones, Científicas y Técnicas (CONICET)-Universidad Nacional del Comahue,

18 8400 Bariloche, Argentina

19 5Instituto de Investigación y Desarrollo en Ingeniería de Procesos, Biotecnología y Energías

20 Alternativas (PROBIEN, CONICET-UNCo), Neuquén, Argentina

21 @Corresponding authors: CTH [email protected] & DL [email protected]

1 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

24 Abstract:

25 S. eubayanus, the wild, cold-tolerant parent of hybrid lager-brewing yeasts, has a

26 complex and understudied natural history. The exploration of this diversity can be used both to

27 develop new brewing applications and to enlighten our understanding of the dynamics of yeast

28 evolution in the wild. Here, we integrate whole genome sequence and phenotypic data of 200 S.

29 eubayanus strains, the largest collection to date. S. eubayanus has a multilayered population

30 structure, consisting of two major populations that are further structured into six subpopulations.

31 Four of these subpopulations are found exclusively in the Patagonian region of South America;

32 one is found predominantly in Patagonia and sparsely in Oceania and North America; and one is

33 specific to the Holarctic ecozone. S. eubayanus is most abundant and genetically diverse in

34 Patagonia, where some locations harbor more genetic diversity than is found outside of South

35 America. All but one subpopulation shows isolation-by-distance, and gene flow between

36 subpopulations is low. However, there are strong signals of ancient and recent outcrossing,

37 including two admixed lineages, one that is sympatric with and one that is mostly isolated from

38 its parental populations. Despite S. eubayanus’ extensive genetic diversity, it has relatively little

39 phenotypic diversity, and all subpopulations performed similarly under most conditions tested.

40 Using our extensive biogeographical data, we constructed a robust model that predicted all

41 known and a handful of additional regions of the globe that are climatically suitable for S.

42 eubayanus, including Europe. We conclude that this industrially relevant species has rich wild

43 diversity with many factors contributing to its complex distribution and biology.

2 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

47 Introduction:

48 In microbial population genomics, the interplay of human association and natural

49 variation is still poorly understood. The genus Saccharomyces is an optimal model to address

50 these questions for eukaryotic microbes, as it contains both partly human-associated species (i.e.

51 Saccharomyces cerevisiae) and mostly wild species (e.g. Saccharomyces paradoxus). These two

52 examples also illustrate the complexity of studying yeast population genomics. Much of S.

53 cerevisiae population structure is admixed, and several lineages show signatures of

54 domestication (Liti et al. 2009; Schacherer et al. 2009; Gallone et al. 2016; Gonçalves et al.

55 2016). In contrast, S. paradoxus is almost exclusively found in the wild and has a population

56 structure that is correlated with geography (Leducq et al. 2014; Eberlein et al. 2019). Pure

57 isolates of their more distant relative Saccharomyces eubayanus have only ever been isolated

58 from wild environments; yet, hybridizations between S. cerevisiae and S. eubayanus were key

59 innovations that enabled cold fermentation and lager brewing (Libkind et al. 2011; Gibson and

60 Liti 2015; Hittinger et al. 2018; Baker et al. 2019). Other hybrids with contributions from S.

61 eubayanus have been isolated from industrial environments (Almeida et al. 2014; Nguyen and

62 Boekhout 2017), indicating that this species has long been playing a role in shaping many

63 fermented products. This association with both natural and domesticated environments makes S.

64 eubayanus an excellent model where both wild diversity and domestication can be investigated.

65 Since the discovery of S. eubayanus in Patagonia (Libkind et al. 2011), this species has

66 received much attention, both for brewing applications and understanding the evolution, ecology,

67 population genomics of the genus Saccharomyces (Sampaio 2018). In the years since its

68 discovery, many new globally distributed isolates have been found (Bing et al. 2014; Peris et al.

69 2014; Rodríguez et al. 2014; Gayevskiy and Goddard 2016; Peris et al. 2016; Eizaguirre et al.

3 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

70 2018). Prior research has suggested that S. eubayanus is most abundant and diverse in the

71 Patagonian region of South America, where there are two major populations (Patagonia

72 A/Population A/PA and Patagonia B/Population B/PB) that recent multilocus data suggested are

73 further divided into five subpopulations (PA-1, PA-2, PB-1, PB-2, and PB-3) (Eizaguirre et al.

74 2018). There are two early-diverging lineages, West China and Sichuan, which were identified

75 through multilocus data (Bing et al. 2014) and whose sequence divergences relative to other

76 strains of S. eubayanus are nearly that of currently recognized species boundaries (Peris et al.

77 2016; Sampaio and Gonçalves 2017; Naseeb et al. 2018). A unique admixed lineage has been

78 found only in North America, which has approximately equal contributions from PA and PB

79 (Peris et al. 2014, 2016). Other isolates from outside Patagonia belong to PB, either the PB-1

80 subpopulation that is also found in Patagonia (Gayevskiy and Goddard 2016; Peris et al. 2016),

81 or a Holarctic-specific subpopulation that includes isolates from Tibet and from North Carolina,

82 USA (Bing et al. 2014; Peris et al. 2016). This Holarctic subpopulation includes the closest

83 known wild relatives of the S. eubayanus subgenomes of lager-brewing yeasts (Bing et al. 2014;

84 Peris et al. 2016).

85 To explore the geographic distribution, ecological niche, and genomic diversity of this

86 industrially relevant species, here, we present an analysis of whole genome sequencing data for

87 200 S. eubayanus strains. This dataset confirms the previously proposed population structure

88 (Peris et al. 2014, 2016; Eizaguirre et al. 2018) and extends the analysis to fully explore genomic

89 diversity. Even though S. eubayanus is genetically diverse and globally distributed, there are not

90 large phenotypic differences between subpopulations. This genomic dataset includes evidence of

91 gene flow and admixture in sympatry, as well as admixture in parapatry or allopatry. While S.

92 eubayanus has a well-differentiated population structure, isolation by distance occurs within

4 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

93 subpopulations that are found globally, as well as within subpopulations restricted to a handful of

94 locations. Much of the genetic diversity is limited to northern Patagonia, but modeling suggests

95 that there are more geographic areas that are climatically suitable for this species, including

96 Europe. S. eubayanus maintains genetic diversity over several dimensions, including multiple

97 high-diversity sympatric populations and a low-diversity widespread invasive lineage. The

98 diversity and dispersal of this eukaryotic microbial species mirror observations in plants and

99 animals, including humans, which shows how biogeographical and evolutionary forces can be

100 shared across organismal sizes, big and small.

101

102 Results:

103 Global and regional S. eubayanus population structure and ecology:

104 To expand on existing data (Libkind et al. 2011; Bing et al. 2014; Peris et al. 2014;

105 Rodríguez et al. 2014; Gayevskiy and Goddard 2016; Peris et al. 2016; Eizaguirre et al. 2018),

106 we sequenced the genomes of 174 additional strains of S. eubayanus, bringing our survey to 200

107 S. eubayanus genomes. This large collection provides the most comprehensive dataset to date for

108 S. eubayanus. We note that the dataset does not contain West China or Sichuan strains (Bing et

109 al. 2014), which were unavailable for study and may constitute a distinct species or subspecies.

110 These strains were globally distributed (Figure 1A), but the majority of our strains were from

111 South America (172 total, 155 newly sequenced here). The next most abundant continent was

112 North America with 26 strains (19 new to this publication). We also analyzed whole genome

113 sequence data for the single strain from New Zealand (Gayevskiy and Goddard 2016) and the

114 single Tibetan isolate with available whole genome sequence data (Bing et al. 2014; Brouwers et

115 al. 2019). The collection sites in South America span from northern Patagonia to Tierra del

5 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

116 Fuego (Figure 1B), while the North American isolates have been sparsely found throughout the

117 continent, including the Canadian province of New Brunswick and the American states of

118 Washington, Wisconsin, North Carolina, and South Carolina (Figure 1C).

119

PA−1 PB−1 PB−3 NoAm B PA−2 PB−2 Holarctic SoAm −40

A C 50

50 45 −45

0 Latitude 35 Latitude −50

Latitude 30

−50 25 −120 −100 −80 −60 −55 Longitude

−100 0 100 200 −75 −70 −65 −60 Longitude Longitude

D 50 PB−2 PA−2 PA−1

PB−3 Holarctic SoAm 50 F PB−3 PB−1 E 0 PA−1 25 50 PB−1 PA−2 PB−2 PB−1 PA−1 0 PB−2 PB−2 PA−2 PA−1 −50 PB−3 0 PB−3 −25 PA−2 Holarctic PB1 Holarctic PC2 = 9.95% Hybrid PC2 = 7.81%

PC2 = 13.96% Hybrid NoAm −50 −50 −100 SoAm Holarctic −75 −150 −100 −50 0 50 −50 0 50 100 NoAm PC1 = 41.56% −150 PC1 = 19.03%

−100 0 100 200 PC1 = 40.06% 120 Figure 1. S. eubayanus distribution and population structure. 121 S. eubayanus has a global distribution and two major populations with six subpopulations. (A) Isolation 122 locations of S. eubayanus strains included in the dataset. For visibility, circle size is not scaled by the 123 number of strains. Subpopulation abundance is shown as pie charts. The Patagonian sampling sites have 124 been collapsed to two locations for clarity. Details of sites and subpopulations found in South America 125 (B) and North America (C) with circle size scaled by the number of strains. (D) Whole genome PCA of S. 126 eubayanus strains and five hybrids with large contributions from S. eubayanus. (E) PCA of just PA. (F) 127 PCA of just PB and hybrid S. eubayanus sub-genomes. Color legends in A and D apply to this and all 128 other figures. 129 130 To determine population structure, we took several approaches, including Principal

131 Component Analysis (PCA) (Jombart 2008), phylogenomic networks (Huson and Bryant 2006),

132 and STRUCTURE-like analyses (Lawson et al. 2012; Raj et al. 2014). All methods showed that

133 S. eubayanus has two large populations that can be further subdivided into a total of six non-

6 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

134 admixed subpopulations and one abundant North American admixed lineage (Figure 1D and

135 Figure S1). We previously described the two major populations, PA and PB-Holarctic (Peris et

136 al. 2014, 2016), as well as the subpopulations PA-1, PA-2, PB-1, PB-2, Holarctic, and the North

137 American admixed lineage (Peris et al. 2016). PB-3 had been suggested by multilocus data

138 (Eizaguirre et al. 2018), and our new analyses confirm this subpopulation with whole genome

139 sequence data. All of the strains isolated from outside of South America belonged to either the

140 previously described North American admixed lineage (NoAm) or one of two PB

141 subpopulations, PB-1 or Holarctic. This dataset included novel PB-1 isolates from the states of

142 Washington (yHRMV83) and North Carolina (yHKB35). Unexpectedly, from this same site in

143 North Carolina, we also obtained new isolates of the NoAm admixed lineage (Figure 1C and

144 Table S1), and we obtained additional new NoAm strains in South Carolina. Together, with the

145 North Carolina strains reported here and previously (Peris et al. 2016), this region near the Blue

146 Ridge Mountains harbors three subpopulations or lineages, PB-1, Holarctic, and NoAm. We

147 were also successful in re-isolating the NoAm lineage from the same Wisconsin site, sampling

148 two years later than what was first reported (Peris et al. 2014) (Table S1), indicating that the

149 NoAm admixed lineage is established, not ephemeral, in this location. Additionally, we found

150 one novel South American strain that was admixed between PA (~45%) and PB (~55%) (Figure

151 1D “SoAm”). This global distribution and the well-differentiated population structure of S.

152 eubayanus is similar to what has been observed in S. paradoxus (Leducq et al. 2014, 2016) and

153 Saccharomyces uvarum (Almeida et al. 2014).

154 S. eubayanus has been isolated from numerous substrates and hosts, and our large dataset

155 afforded us the power to analyze host and substrate association by subpopulation. We found that

156 PA-2 was associated with the seeds of Araucaria araucana (45.71% of isolates, p-val = 6.11E-

7 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

157 07, F-statistic = 15.29). Interestingly, while PB-1 was the most frequently isolated subpopulation

158 (34% of isolates), it has never been isolated from A. araucana seeds. Instead, PB-1 was

159 associated with Nothofagus antarctica (52.31% of isolates, p-val = 0.017, F-statistic = 3.10). PB-

160 1 was also the subpopulation isolated the most from Nothofagus dombeyi (75% of isolates from

161 this tree species), which is a common host of S. uvarum (Libkind et al. 2011; Eizaguirre et al.

162 2018). PB-2 was positively associated with Nothofagus pumilio (36.59% of isolates, p-val = 9.60

163 E-04, F-statistic = 6.59), which could be an ecological factor keeping PB-2 partly isolated from

164 its sympatric subpopulations, PA-2 and PB-1 (Figure 1C). PB-3 was associated with the fungal

165 parasite Cyttaria darwinii (14.29% of isolates, p-val = 0.039, F-statistic = 25.34 ) and

166 Nothofagus betuloides (28.57% of isolates, p-val = 5.02E-06, F-statistic = 60.35), which is only

167 found in southern Patagonia and is vicariant with N. dombeyi, a host of PB-1. PB-3 was

168 frequently isolated in southern Patagonia (49% of southern isolates) (Eizaguirre et al. 2018), and

169 its association with a southern-distributed tree species could play a role in its geographic range

170 and genetic isolation from the northern subpopulations. Neither Nothofagus nor A. araucana are

171 native to North America, and we found that our North American isolates were from multiple

172 diverse plant hosts, including Juniperus virginiana, Diospyros virginiana, Cedrus sp., and Pinus

173 sp. (Table S1), as well as from both soil and bark samples. In Patagonia, S. eubayanus has been

174 isolated from exotic Quercus trees (Eizaguirre et al. 2018), so even though Nothofagus and A.

175 araucana are common hosts, S. eubayanus can be found on a variety of hosts and substrates.

176 These observed differences in host and substrate could be playing a role in the maintenance of its

177 population structure, especially in sympatric regions of Patagonia.

178

179 All subpopulations grow at freezing temperatures and on diverse carbon sources:

8 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

180 S. eubayanus comes from a wide range of environments, so we tested if there were

181 phenotypic differences between these subpopulations. We measured growth rates on several

182 carbon sources and stress responses for a large subset of these strains (190) and 26 lager-brewing

183 strains (Figure 2 and Figure S2). Lager-brewing strains grew faster on maltotriose than all

184 subpopulations (p-val < 0.05, Figure 2A), which is consistent with this sugar being one of the

185 most abundant in brewing wort but rare in nature (Salema-Oom et al. 2005). The Holarctic

186 subpopulation grew slower on glucose and maltose compared to all other subpopulations (p-val <

187 0.05, Figure 2A, Table S2). Overall, the admixed NoAm lineage performed better than PB-1 (p-

188 val = 0.038, Figure 2A), but there was no interaction with carbon source. Therefore, the admixed

189 lineage’s robustness in many conditions could play a role in its success in far-flung North

190 American sites where no pure PA or PB strains have ever been found.

A GR B GR PA−2 0.06 Lager NoAm 0.15 PA−1 Holarctic PA−2 PAPB−−22 0.060.05 Lager NoAmPB−3 NoAm 0.15 PA−1 Holarctic 0.1 PB−1 0.050.04 PB−2 PBLager−2 PAPB−−23 PBHolarctic−3 NoAm Raffinose Maltotriose Glycerol Ethanol Glucose Maltose Galactose 0.03 PB−1 PB−1 0.04 PB−2 0.1 Lager PA−1 0.05 Holarctic 0.02 20C 10C 0C 4C PB−3 Raffinose Maltotriose Glycerol Ethanol Glucose Maltose Galactose 0.03 PB−1 PA−1 0.05 0.020.01 20C 10C 0C 4C

0.01 191 192 Figure 2. Phenotypic differences. 193 (A) Heat map of mean of maximum growth rate (change in OD/hour) (GR) on different carbon sources by 194 subpopulation. Warmer colors designate faster growth. (B) Heat map of log10 normalized growth at 195 different temperatures by subpopulation. 196 197 Since S. eubayanus’ contribution to the cold-adaptation of hybrid brewing strains is well

198 established (Libkind et al. 2011; Gibson et al. 2013; Baker et al. 2019), we measured growth at

199 0°C, 4°C, 10°C, and 20°C. All subpopulations grew at temperatures as low as 0°C (Figure 2B

9 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

200 and Figure S2), and all S. eubayanus subpopulations outperformed lager-brewing yeasts (p <

201 0.05). Within pure S. eubayanus, there were no temperature by subpopulation interactions,

202 indicating that no subpopulation is more cryotolerant than any other subpopulation. In summary,

203 we found that all strains that we tested grew similarly in many environments, and despite the

204 large amount of genotypic diversity observed for this species, we observed much less phenotypic

205 diversity (Figure 2).

206

207 Subpopulations are well differentiated:

208 The mating strategies and life cycle of Saccharomyces, with intratetrad mating and

209 haploselfing, often lead to homozygous diploid individuals (Hittinger 2013). Nonetheless, in S.

210 cerevisiae, many industrial strains are highly heterozygous (Gallone et al. 2016; Gonçalves et al.

211 2016; Peter et al. 2018). Here, we analyzed genome-wide heterozygosity in our collection of 200

212 strains and found only one individual with more than 20,000 heterozygous SNPs (Figure S3).

213 When we phased highly heterozygous regions of its genome and analyzed the two phases

214 separately, we found that both phases grouped within PB-1 (Figure S3C). Thus, while this strain

215 is highly heterozygous, it has contributions from only one subpopulation.

216 This large collection of strains is a powerful resource to explore natural variation and

217 population demography in a wild microbe, so we analyzed several common population genomic

218 statistics in 50-kbp windows across the genome. We found that diversity was similar between

219 subpopulations (Figure S4A). We also calculated Tajima’s D and found that the genome-wide

220 mean was zero or negative for each subpopulation (Figure S4B), which could be indicative of

221 population expansions. In particular, the most numerous and widespread subpopulation, PB-1,

10 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

222 had the most negative and consistent Tajima’s D, suggesting a recent population expansion is

223 especially likely in this case.

PA−1 B A PB−3 0.6

0.78 0.69 0.67 0.72 0.43 PB−2 Subpop 0.51 PA−1 0.73 0.4 PB−1 0.64 ● PA−2

0.70 SoAm 2 r PB−1 0.59 0.78 PB−2 PA−2 PB−3 Hol Seub

0.2

0.64 0.80

0.0 ● 0KB 30KB 60KB 90KB 120KB 150KB 180KB 210KB 240KB 270KB 224 NoAm SNP distance 225 Figure 3. Population genomic parameters. 226 (A) Network built with pairwise FST values < 0.8 between each subpopulation. FST values are printed and 227 correspond to line thickness, where lower values are thicker. Circle sizes correspond to genetic diversity. 228 (B) LD decay for each subpopulation (colors) and the species in whole (black). 229 230 For the non-admixed lineages, genome-wide average FST was consistently high across the

231 genome (Figure S4C). In pairwise comparisons of FST, PB-1 had the lowest values of any

232 subpopulation (Figure 3A, Figure S4D). These pairwise comparisons also showed that, within

233 each population, there has been some gene flow between subpopulations, even though the

234 subpopulations were generally well differentiated. Linkage disequilibrium (LD) decay indicated

235 low recombination in these wild subpopulations (Figure 3B), with variability between

236 subpopulations. For the species as a whole, LD decayed to one-half at about 5 kbp, which is

237 somewhat higher than the 500bp - 3kbp observed in S. cerevisiae (Liti et al. 2009; Schacherer et

238 al. 2009; Peter et al. 2018) and lower than the 9 kbp observed in S. paradoxus (Liti et al. 2009),

239 indicating that there is less mating, outcrossing, and/or recombination in this wild species than S.

240 cerevisiae and more than in S. paradoxus.

11 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

241

242 Recent admixture and historical gene flow between populations:

243 We previously reported the existence of 7 strains of an admixed lineage in Wisconsin,

244 USA, and New Brunswick, Canada (Peris et al. 2014, 2016). Here, we present 14 additional

245 isolates of this same admixed lineage. These new isolates were from the same site in Wisconsin,

246 as well as two new locations in North Carolina and South Carolina (Table S1). Strikingly, all 21

247 strains shared the exact same genome-wide ancestry profile (Figure 4A), indicating that they all

248 descended from the same outcrossing event between the two main populations of S. eubayanus.

249 These admixed strains were differentiated by 571 SNPs, which also delineated these strains

250 geographically (Figure 4B). Pairwise diversity and FST comparisons across the genomes suggest

251 that the PA parent came from the PA-2 subpopulation (Figure 4C and Figure S5A) and that the

252 PB parent was from the PB-1 subpopulation (Figure 4D and Figure S5A).

A yHDPN421 0.1 B 0.1 E SoAm yHKS 694 yHKB201 yHQL1481 North Wisconsin, yHKS 689 yHKB251 yHDPN422 yHKS 576 Carolina, USA 5 )

USA yHKS 212 m

yHKS 810 A yHKS 210 o S

yHDPN423 yHQL1621 - yHKS 211 yHKS 812 B d

yHKB210 m 0 A o

yHDPN424 S yHDP N422 - A

yHDP N424 d ( yHDP N423yHDP N421 2 g o yHKB201 l New −5 yHKB246 Brunswick, yHKB242 yHKB210 yHKB268 Canada South Carolina, USA yHKB218 I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI yHKB218 NoAm Genome Position yHKB242 C F

) yHKB246 1.5 1.5 m A o N

- yHKB251 B d 1.0 Subpop 1.0 Subpop m yHKB268 A PA−1 PA−1 o

N PA−2 PA−2 - A yHKS210 d

( 0.5 0.5 2 Pairwise Divergence (%) Divergence Pairwise (%) Divergence Pairwise g o

l yHKS211

yHKS212 0.0 0.0

I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI yHKS576 Genome Position Genome Position D G yHKS689 1.5 1.5 yHKS694

Subpop Subpop yHKS810 1.0 1.0 PB−1 PB−1 PB−2 PB−2 yHKS812 PB−3 PB−3

0.5 0.5 Pairwise Divergence (%) Divergence Pairwise yHQL1481 (%) Divergence Pairwise

yHQL1621 0.0 0.0 I II III IV VIVIIVIII IX V X XI XII XIIIXIVXVXVI I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI 253 Genome Position Genome Position Genome Position 254 Figure 4. Genomic ancestries of NoAm and SoAm admixed lineages.

12 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

255 (A) For all 21 NoAm admixed strains, log2 ratio of the minimum PB-NoAm pairwise nucleotide sequence 256 divergence (dB-NoAm) and the minimum PA-NoAm pairwise nucleotide sequence divergence (dA- 257 NoAm) in 50-kbp windows. Colors and log2 < 0 or > 0 indicate that part of the genome is more closely 258 related to PA or PB, respectively. (B) Neighbor-Net phylogenetic network reconstructed with the 571 259 SNPs that differentiate the NoAm strains. The scale bar represents the number of substitutions per site. 260 Collection location is noted. (C) Pairwise nucleotide sequence divergence of the NoAm strain yHKS210 261 compared to strains from the PA-1 and PA-2 subpopulations of PA in 50-kbp windows. (D) Pairwise 262 nucleotide sequence divergence of the NoAm strain yHKS210 compared to strains from the PB-1, PB-2, 263 and PB-3 subpopulations of PB in 50-kbp windows. (E) log2 ratio of the minimum PB-SoAm pairwise 264 nucleotide sequence divergence (dB-SoAm) and the minimum PA-SoAm pairwise nucleotide sequence 265 divergence (dA-SoAm) in 50-kbp windows. Colors and log2 < 0 or > 0 indicate that part of the genome is 266 more closely related to PA or PB, respectively. (F) Pairwise nucleotide sequence divergence of the SoAm 267 strain compared to strains from the PA-1 and PA-2 subpopulations of PA in 50-kbp windows. (G) 268 Pairwise nucleotide sequence divergence of the SoAm strain compared to strains of the PB-1, PB-2, and 269 PB-3 subpopulations of PB in 50-kbp windows. 270 271 Here, we report a second instance of recent outcrossing between PA and PB. One other

272 strain with fairly equal contributions from the two major populations, PA (~45%) and PB

273 (~55%) (Figure 4E), was isolated from the eastern side of Nahuel Huapi National Park, an area

274 that is sympatric for all subpopulations found in South America. This strain had a complex

275 ancestry, where both PA-1 and PA-2 contributed to the PA portions of its genome (Figure 4F and

276 Figure S5B), indicating that its PA parent was already admixed between PA-1 and PA-2. As with

277 the NoAm admixed strains, the PB parent was from the PB-1 subpopulation (Figure 4G and

278 Figure S5B). Together, these two admixed lineages show that outcrossing occurs between the

279 two major populations, and that admixture and gene flow are likely ongoing within sympatric

280 regions of South America.

281 We also found examples of smaller tracts of admixture between PA and PB that were

282 detectable as 2-12% contributions. These introgressed strains included the taxonomic type strain

283 of S. eubayanus (CBS12357T), whose genome sequence was mostly inferred to be from PB-1,

284 but it had a ~4% contribution from PA-1 (Figure S6). We found several other examples of

285 admixture between PA and PB, as well as admixture between subpopulations of PA or of PB

286 (Table S3). Notably, the PB contributions were usually from PB-1, the subpopulation with the

13 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

287 largest range, most hosts, and strongest signature of population expansion, factors that would

288 tend to make contact with other subpopulations more likely.

289 In our collection of 200 strains, we observed nuclear genome contributions from S.

290 uvarum in four strains. These four strains all shared the same introgression of ~150-kbp on

291 chromosome XIV (Figure S7A&B). When we analyzed the portion of the genome contributed by

292 S. eubayanus, we found that these strains were all embedded in the PB-1 subpopulation (Figure

293 S7C). Analysis of the 150-kbp region from S. uvarum indicated that the closest S. uvarum

294 population related to these introgressed strains was SA-B (Figure S7D), a population restricted to

295 South America that has not previously been found to contribute to any known interspecies

296 hybrids (Almeida et al. 2014). These strains thus represent an independent hybridization event

297 between South American lineages of these two sister species that is not related to any known

298 hybridization events among industrial strains (Almeida et al. 2014). These strains show that S.

299 eubayanus and S. uvarum can and do hybridize in the wild, but the limited number (n=4) of

300 introgressed strains, small introgression size (150-kbp), and shared breakpoints suggest that the

301 persistence of hybrids in the wild is rare.

302

303 Northern Patagonia is a diversity hot spot:

304 Patagonia harbors the most genetic diversity of S. eubayanus in our dataset, and four

305 subpopulations were found only there: PA-1, PA-2, PB-2, and PB-3 (Figure 1A and 4A).

306 Therefore, we examined the genetic diversity and range distributions of the isolates from South

307 America more closely. Nahuel Huapi National Park yielded isolates from all five subpopulations

308 found in South America, was the only place where PA-1 was found, and was the location where

309 the SoAm admixed strain was isolated (Figure 5A & B). All five sub-populations were found

14 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

310 north of 43°S, an important boundary during the last glaciation period that affects many

311 organisms (Mathiasen and Premoli 2010; Premoli et al. 2010; Quiroga and Premoli 2010).

312 Species-wide, there was more genetic diversity north of this boundary (Figure 5B). In contrast,

313 only PB-1 and PB-3 were found south of 43°S, with both distributions reaching Tierra del Fuego.

314 The southernmost strains were primarily PB-3 (89.7%), but they included two highly admixed

315 PB-1 × PB-3 strains (Table S1 & S3).

A B C D PA−1 PA−2 PB−1 PB−2 PB−3 All Non-South American 0.20 ● PA−1 PA−2 NoAm PB−1 PB−2 PB−3 Hol All North American Subpopulation

● PA−1

● PA−2

● PB−1 −40 −40 ● ● PB−2 0.1000 0.15 ● ● ● PB−3 Nucleotide ● NoAm Diversity (%) 0.0 Population Geographic 0.1 ● ● Size 0.2 −45 −45 100000 0.0100 0.3 0.10 Latitude

0.4 200000 Latitude ● ● Nucleotide Diversity (%) Nucleotide Diversity

0.5

300000 0.0010

−50 −50 (%)) log(Nucleotide Diversity 0.05

400000

0.0001

−55 −55 0.00 ● ● p-val=0.564 p-val=0.0003 p-val=0.0001 p-val=0.0001 p-val=0.0001 p-val=0.0001 p-val=0.0001

PA−1 PA−2 PB−1 PB−2 PB−3 SoAm SA World Area Distance Sites Sites (10 Km^2) (Km) 0 10 20 0 200 400 0 500 10001500 0 5000 10000 0 100 200 0 500 1000 1500 0 5000 10000 −75 −70 −65 −60 Distance (Km) 316 Longitude 317 Figure 5. South American genomic diversity versus range, diversity by area, and isolation by distance. 318 (A) Range and genomic diversity of South American sampling sites. Circle sizes correspond to nucleotide 319 diversity of all strains from that site, and pie proportions correspond to each subpopulation’s contribution 320 to � at each site. Latitudinal range of each subpopulation is shown to the right. (B) Nucleotide diversity 321 by subpopulation by sampling site, where larger and darker circles indicate more diversity. “SA Sites” in 322 gray show the diversity of all strains found in each South American (SA) site. “World Sites” in darker 323 gray show the nucleotide diversity of all North American or non-South American strains, regardless of 324 subpopulation, compared to South American strains south or north of 43°S, aligned to mean latitude of all 325 strains included in the analysis. (C) Correlation of nucleotide diversity and the area or distance a 326 subpopulation covers. The y-axis shows the nucleotide diversity of each subpopulation, and circle sizes 327 correspond to the geographic sizes of the subpopulations on a log10 scale. Note that PA-1 (dark red) is as 328 diverse as PB-3 (dark blue) but encompasses a smaller area. (D) log10(pairwise nucleotide diversity) 329 correlated with distance between strains, which demonstrates isolation by distance. Note that y-axes are 330 all scaled the same but not the x-axes. Holarctic includes the S. eubayanus sub-genome of two lager- 331 brewing strains. Figure S8A shows the individual plots for the NoAm lineage. Figure S8B shows the 332 individual plot of PB-1. 333 334 Despite the limited geographic range of some subpopulations, their genetic diversity was

335 high, and this diversity often did not scale with the geographic area over which they were found

336 (Figure 5C). The widespread distribution of some subpopulations led us to question if there was

337 isolation by distance within a subpopulation (Figure 5D). We used pairwise measures of

15 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

338 diversity and geographic distance between each strain and conducted Mantel tests for each

339 subpopulation. All subpopulations showed significant isolation by distance (Table S4), except

340 PA-1, likely because it had the smallest geographic range (25 km). Even the Mantel test for the

341 least diverse lineage, NoAm, was highly significant (p-val = 0.0001, R2= 0.106), indicating that

342 each location has been evolving independently after their recent shared outcrossing and dispersal

343 event. Through these pairwise analyses, we also detected two strains from Cerro Ñielol, Chile,

344 that were unusually genetically divergent from the rest of PB-1 and could potentially be a novel

345 lineage (Figure S8).

346

347 Additional global regions are climatically suitable:

348 The sparse but global distribution of S. eubayanus raises questions about whether other

349 areas of the world could be suitable for this species. We used the maxent environmental niche

350 modeling algorithm implemented in Wallace (Kass et al. 2018) to model the global climatic

351 suitability for S. eubayanus, using GPS coordinates of all known S. eubayanus strains published

352 here and estimates of coordinates for the East Asian isolates (Bing et al. 2014). These niche

353 models were built using the WorldClim Bioclims, which are based on monthly temperature and

354 rainfall measures, reflecting both annual and seasonal trends, as well as extremes, such as the

355 hottest and coldest quarters. How climatic variables affect yeast distributions is an understudied

356 area, and building these models allowed us a novel way to explore climatic suitability.

357 Using all known locations of isolation (Figure 6), we found that the best model

358 (AIC=2122.4) accurately delineated the known distribution along the Patagonian Andes. In

359 North America, the strains from the Olympic Mountains of Washington state and the Blue Ridge

360 region of North Carolina fell within the predicted areas, and interestingly, these sites had yielded

16 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

361 pure PB-1 and Holarctic strains. In contrast, some of the NoAm admixed strains were found in

362 regions that were on the border of suitability in this model (New Brunswick and Wisconsin). In

363 Asia, the model predicts further suitable regions along the Himalayas that are west of known

364 locations.

Predicted Climatic Suitability

1.0

0.5 60°N 0.0

40°N I

20°N IV

Latitude 0°

Ancestor 20°S

III 40°S

150°W 100°W 50°W 0° 50°E 100°E 150°E 365 Longitude

366 Figure 6. Predicted climatic suitability of S. eubayanus. 367 Minimum training presence (light green) and 10th percentile training presence (dark green) based on a 368 model that includes all known S. eubayanus isolations, as well as a scenario of dispersal and 369 diversification out of Patagonia (inset and arrows). Black arrows signify diversification events, dotted 370 lines are diversification events where the population is not found in Patagonia, and colored arrows are 371 migration events for the lineage of matching color. Roman numerals order the potential migration events. 372 S. eubayanus has not been found in the wild in Europe, but it has contributed to fermentation hybrids, 373 such as lager yeasts. This scenario proposes that the last common ancestor of PA and PB-Holarctic 374 bifurcated into PA (red) and PB-Holarctic (blue), which further radiated into PA-1 (dark red), PA-2 (light 375 red), PB-1 (blue), PB-2 (lighter blue), PB-3 (dark blue), and Holarctic (very light blue). At least four 376 migration events are needed to explain the locations where S. eubayanus has been found. I. The Holarctic 377 subpopulation was drawn from the PB-Holarctic gene pool and colonized the Holarctic ecozone. II. PB-1 378 colonized the Pacific Rim, including New Zealand and Washington state, USA. III. An independent 379 dispersal event brought PB-1 to North Carolina, USA. IV. Outcrossing between PA-2 and PB-1 gave rise 380 to a low-diversity admixed linage that has recently invaded a large swath of North America. 381 382 The uneven global distribution of S. eubayanus led us to test if models were robust to

383 being built only with the South American locations or only with the non-South American

384 locations (Figure S9). Remarkably, with just the South American isolates, the model

385 (AIC=1327.32) accurately predicted the locations of the non-South American isolates (Figure

17 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

386 S9A). Even the model built from the limited number of isolates from outside South America

387 (AIC=558.58) still performed reasonably well, identifying the regions in Patagonia along the

388 Andes where S. eubayanus has been found (Figure S9B). Collectively, these models suggest that

389 climatic modeling can predict other suitable regions for eukaryotic microbes. These approaches

390 could be used to direct future sampling efforts or applied to other microbes to gain further insight

391 into microbial ecology.

392 Notably, all models agree that Europe is climatically a prime location for S. eubayanus

393 (Figure S9C), but no pure isolates have ever been found there, only hybrids with S. uvarum, S.

394 cerevisiae, or hybrids with even more parents (Almeida et al. 2014). These hybrids with complex

395 ancestries have been found in numerous fermentation environments, suggesting that pure S.

396 eubayanus once existed, or still exists at low abundance or in obscure locations, in Europe. Thus,

397 the lack of wild isolates from sampling efforts in Europe remains a complex puzzle.

398

399 Discussion:

400 Here, we integrated genomic, geographic, and phenotypic data for 200 strains of S.

401 eubayanus, the largest collection to date, to gain insight into its world-wide distribution, climatic

402 suitability, and population structure. All the strains belong to the two major populations

403 previously described (Peris et al. 2014, 2016), but with the extended dataset, we were able to

404 define considerable additional structure, consisting of six subpopulations and two admixed

405 lineages. These subpopulations have high genetic diversity, high FST, and long LD decay; all

406 measures indicative of large and partly isolated populations undergoing limited gene flow.

407 Despite this high genetic diversity, there was relatively little phenotypic differentiation between

408 subpopulations. This dichotomy between large genetic diversity and limited phenotypic

18 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

409 differentiation hints at a complex demographic history where genetically differentiated

410 subpopulations are minimally phenotypically differentiated and grow well in a wide range of

411 environments.

412 Despite the strong population structure, we also observed considerable evidence of

413 admixture and gene flow. The two recently admixed lineages had nearly equal contributions

414 from the two major populations, but they were the result of independent outcrossing events. The

415 SoAm admixed strain was isolated from a hotspot of diversity and contains contributions from

416 three subpopulations. The NoAm admixed lineage has spread across at least four distant

417 locations, but all strains descended from the same outcrossing event. Since PA has only been

418 isolated in South America, it is intriguing that the NoAm admixed lineage has been successful in

419 so many locations throughout North America. The success of this lineage could be partially

420 explained by its equal or better performance in many environments in comparison to its parental

421 populations (Fig. 2), perhaps contributing to its invasion of several new locations. Several other

422 Patagonian strains also revealed more modest degrees of gene flow between PA and PB. Finally,

423 we characterized a shared nuclear introgression from S. uvarum into four Patagonian strains of S.

424 eubayanus, demonstrating that hybridization and backcrossing between these sister species has

425 occurred in the wild in South America.

426 S. eubayanus has a paradoxical biogeographical distribution; it is abundant in Patagonia,

427 but it is sparsely found elsewhere with far-flung isolates from North America, Asia, and Oceania.

428 Most subpopulations displayed isolation by distance, but genetic diversity only scaled with

429 geographic range to a limited extent. In Patagonia, some sampling sites harbor more genetic

430 diversity than all non-Patagonian locations together (Figure 5B). Although we found the most

431 genetic diversity and largest number of subpopulations north of 43°S, the pattern of genetic

19 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

432 diversity appears to be reversed on the west side of the Andes, at least for the PB-1

433 subpopulation (Nespolo et al. 2019). This discrepancy could be due to differences in how glacial

434 refugia were distributed (Sérsic et al. 2011) and limitations on gene flow between the east and

435 west sides of the Andes. Together, the levels of S. eubayanus genetic diversity found within

436 Patagonia, as well as the restriction of four subpopulations to Patagonia, suggest that Patagonia is

437 the origin of most of the diversity of S. eubayanus, likely including the last common ancestor of

438 the PA and PB-Holarctic populations.

439 The simplest scenario to explain the current distribution and diversity of S. eubayanus is

440 a series of radiations in Patagonia, followed by a handful of out-of-Patagonia migration events

441 (Figure 6). Under this model, PA and PB would have bifurcated in Patagonia, possibly in

442 separate glacial refugia. The oldest migration event would have been the dispersal of the ancestor

443 of the Holarctic subpopulation, drawn from the PB gene pool, to the Northern Hemisphere.

444 Multiple more recent migration events could have resulted in the few PB-1 strains found in New

445 Zealand and the USA. The New Zealand and Washington state strains cluster phylogenetically

446 and could have diversified from the same migration event from Patagonia into the Pacific Rim.

447 The PB-1 strain from North Carolina (yHKB35) is genetically more similar to PB-1 strains from

448 Patagonia, suggesting it arrived in the Northern Hemisphere independently of the Pacific Rim

449 strains. Finally, the NoAm admixed strains are likely the descendants of a single, and relatively

450 recent, out-of-Patagonia dispersal. Given that PA appears to be restricted to northern Patagonia,

451 this region could have been where the hybridization leading to the NoAm lineage occurred.

452 While the dispersal vector that brought this admixed lineage to North America is unknown, its

453 far-flung distribution and low diversity show that it has rapidly succeeded by invading new

454 environments.

20 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

455 Other more complex scenarios could conceivably explain the limited number of strains

456 found outside of Patagonia. For example, PA and PB could represent sequential colonizations of

457 Patagonia from the Northern Hemisphere. Under this model, PA would have arrived first and

458 would then have been restricted to northern Patagonia by competition with the later arrival of

459 PB. The Holarctic subpopulation could be interpreted as remnants of the PB population that did

460 not migrate to Patagonia; but the PB-1 strains from the Northern Hemisphere, especially

461 yHKB35, seem far more likely to have been drawn from a Patagonian gene pool than the other

462 way around. Furthermore, the structuring of the PA-1 and the PA-2 subpopulations and of the

463 PB-1, PB-2, and PB-3 subpopulations are particularly challenging to rectify with models that do

464 not allow for diversification within South America. Even more complex scenarios remain

465 possible, and more sampling and isolation will be required to fully elucidate the distribution of

466 this elusive species and more conclusively reject potential biogeographical models.

467 S. eubayanus has a strikingly parallel population structure and genetic diversity to its

468 sister species S. uvarum (Almeida et al. 2014; Peris et al. 2016). Both species are abundant and

469 diverse in Patagonia but can be found globally. Both have early diverging lineages, found in Asia

470 or Australasia, that border on being considered novel species. In South America, both have two

471 major populations, where one of these populations is restricted to northern Patagonia (north of

472 43°S). However, a major difference between the distribution of these species is that pure strains

473 of S. uvarum have been found in Europe. Many dimensions of biodiversity could be interacting

474 to bound the distribution and population structure of both S. eubayanus and S. uvarum. In

475 particular, we know very little about local ecology, including the biotic community and

476 availability of abiotic resources on a microbial scale, but these factors likely all influence

477 microbial success. We show here that substrate and host association vary between

21 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

478 subpopulations. In Patagonia, S. eubayanus and S. uvarum are commonly associated with

479 Nothofagus, where N. dombeyi is the preferred host of S. uvarum (Libkind et al. 2011; Eizaguirre

480 et al. 2018). Therefore, niche partitioning of host trees could be playing role in the persistence of

481 these species in sympatry in Patagonia. However, in locations where Nothofagus is not found and

482 there are perhaps fewer hosts, competitive exclusion between the sister species S. eubayanus and

483 S. uvarum, could influence distribution. Competition for a narrower set of hosts could potentially

484 explain why only S. uvarum has been found in Europe as pure strains, while S. eubayanus has

485 not. A second factor influencing distribution and population structure could be dispersal. Yeasts

486 could migrate via many avenues, such as wind, insect, bird, or other animals (Francesca et al.

487 2012, 2014; Stefanini et al. 2012; Gillespie et al. 2012). Human mediated-dispersal has been

488 inferred for the S. cerevisiae Wine and Beer lineages and for the S. paradoxus European/SpA

489 lineage (Gallone et al. 2016; Gonçalves et al. 2016; Leducq et al. 2014; Kuehne et al. 2007). A

490 third bounding factor could be a region’s historical climate. Glacial refugia act as reservoirs of

491 isolated genetic diversity that allow expansion into new areas after glacial retreat (Stewart and

492 Lister 2001). 43°S is a significant geographic boundary due to past geological and climatic

493 variables (Mathiasen and Premoli 2010; Eizaguirre et al. 2018), and many other species and

494 genera show a distinction between their northern and southern counterparts, including

495 Nothofagus (Mathiasen and Premoli 2010; Premoli et al. 2012). S. eubayanus and S. uvarum

496 diversities are also strongly affected by the 43°S boundary (Almeida et al. 2014; Eizaguirre et al.

497 2018), and it seems likely that the microbes experienced some of the same glaciation effects as

498 their hosts. The strong correlation of S. eubayanus and S. uvarum population structures with

499 43°S further implies a longstanding and intimate association with Patagonia.

22 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

500 The sparse global distribution and complex patterns of genetic diversity continue to raise

501 questions about the niche and potential range of S. eubayanus. Our climatic modeling suggests

502 that parts of Europe would be ideal for S. eubayanus. Despite extensive sampling efforts, S.

503 eubayanus has never been isolated in Europe (Sampaio 2018). However, recent environmental

504 sequencing of the fungal specific ITS1 region hinted that S. eubayanus may exist in the wild in

505 Europe (Alsammar et al. 2019). Considerable caution is warranted in interpreting this result

506 because the rDNA locus quickly fixes to one parent’s allele in interspecies hybrids, there is only

507 a single ITS1 SNP between S. uvarum and S. eubayanus, and the dataset contained very few

508 reads that mapped to S. eubayanus. Still, the prevalence of hybrids with contributions from the

509 Holarctic lineage of S. eubayanus found in Europe (Peris et al. 2016) suggests that the Holarctic

510 lineage exists in Europe, or at least existed historically, allowing it to contribute to many

511 independent hybridization events.

512 The patterns of radiation and dispersal observed here mirror the dynamics of evolution

513 found in other organisms (Czekanski-Moir and Rundell 2019), including humans (Nielsen et al.

514 2017). S. eubayanus and humans harbor diverse and structured populations in sub-Saharan

515 Africa and Patagonia, respectively. In these endemic regions, both species show signals of

516 ancient and recent admixture between these structured populations. Both species have

517 successfully colonized wide swaths of the globe, with the consequence of repeated bottlenecks in

518 genetic diversity. While anatomically modern humans underwent a single major out-of-Africa

519 migration that led to the peopling of the world (Nielsen et al. 2017), S. eubayanus has

520 experienced several migration events from different populations that have led to more punctate

521 global distribution. For both species, intraspecific admixture and interspecific hybridization

522 appear to have played adaptive roles in the success of colonizing these new locations. In humans,

23 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

523 introgressions from past hybridizations with both Neanderthals and Denisovans underlie many

524 adaptive traits (Racimo et al. 2015), while the cold fermentation of lager-brewing would not be

525 possible without the cryotolerance of S. eubayanus and the aggressive fermentation of

526 domesticated ale strains of S. cerevisiae (Gibson and Liti 2015). These parallels illustrate how

527 the biogeographical and evolutionary dynamics observed in plants and animals also shape

528 microbial diversity. As yeast ecology and population genomics (Marsit et al. 2017; Yurkov

529 2017) move beyond the Baas-Becking “Everything is everywhere” hypothesis of microbial

530 ecology (Baas-Becking 1934; de Wit and Bouvier 2006), the rich dynamics of natural diversity

531 that is hidden in the soil at our feet is being uncovered.

532

533 Methods:

534 Wild strain isolations

535 All South American isolates were sampled, isolated, and identified as described

536 previously (Libkind et al. 2011; Eizaguirre et al. 2018). North American isolates new to this

537 publication were from soil or bark samples from the American states of Washington, Wisconsin,

538 North Carolina, and South Carolina (Table S1). Strain enrichment and isolation was done as

539 previously described (Sylvester et al. 2015; Peris et al. 2016, 2014), with a few exceptions in

540 temperature and carbon source of isolation (Table S1). Specifically, two strains were isolated at

541 4°C, eight strains were isolated at room temperature, and six strains were isolated on a non-

542 glucose carbon source: three in galactose, two in sucrose, and one in maltose (Table S1).

543

544 Whole genome sequencing and SNP-calling

24 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

545 Whole genome sequencing was completed with Illumina paired-end reads as described

546 previously (Peris et al. 2016; Shen et al. 2018). Reads were aligned to the reference genome

547 (Baker et al. 2015), SNPs were called, masked for low coverage, and retained for downstream

548 analysis as described previously (Peris et al. 2016). One strain, yHCT75, had more than 20,000

549 heterozygous SNPs called. This strain was pseudo-phased using read-backed phasing in GATK

550 (McKenna et al. 2010) and split into two phases. Short-read data is deposited in the NCBI Short

551 Read Archive under PRJNA555221.

552

553 Population genomic analyses:

554 Population structure was defined using several approaches: fastSTRUCTURE (Raj et al.

555 2014), fineSTRUCTURE (Lawson et al. 2012), SplitsTree v4 (Huson and Bryant 2006),

556 and Principal Component Analysis with the adegenet package in R (Jombart 2008).

557 fineSTRUCTURE analysis was completed using all strains and 11994 SNPs. The

558 SplitsTree network was built with this same set of strains and SNPs. fastStructure

559 analysis was completed with as subsample of 5 NoAm strains and 150165 SNPs. We tested K=1

560 through K=10 and selected K=6 using the “chooseK.py” command in fastSTRUCTURE. All

561 calculations of pairwise divergence, FST, and Tajima’s D for subpopulations were computed

562 using the R package PopGenome (Pfeifer and Wittelsbuerger 2015) in windows of 50-kbp.

563 Pairwise divergence between strains was calculated across the whole genome using PopGenome.

564 LD was calculated using PopLDdecay (Zhang et al. 2019). Geographic area and distance of

565 subpopulations was calculated using the geosphere package in R (Hijmans et al. 2019). The

566 Mantel tests were completed using ade4 package of R (Dray and Dufour 2007). The FST network

567 was built with iGraph in R (Csardi and Nepusz 2006).

25 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

568

569 Niche projection with Wallace:

570 Climatic modeling of S. eubayanus was completed using the R package Wallace (Kass et

571 al. 2018). Three sets of occurrence data were tested: one that included only GPS coordinates for

572 strains from South America, one that included only non-South American isolates, and one that

573 included all known isolates (Table S1). We could use exact GPS coordinates for most strains,

574 except for the strains from East Asia, where we estimated the locations (Bing et al. 2014).

575 WorldClim bioclimatic variables were obtained at a resolution of 2.5 arcmin. The background

576 extent was set to “Minimum convex polygon” with a 0.5-degree buffer distance and 10,000

577 background points were sampled. We used block spatial partitioning. The model was built using

578 the Maxent algorithm, using the feature classes: L (linear), LQ (linear quadradic), H (Hinge),

579 LQH, and LQHP (Linear Quadradic Hinge Product) with 1-3 regularization multipliers and the

580 multiplier step value set to 1. The model was chosen based on the Akaike Information Criterion

581 (AIC) score (Table S5). The best models were then projected to the all continents, except

582 Antarctica.

583

584 Phenotyping:

585 Strains were first revived in Yeast Peptone Dextrose (YPD) and grown for 3 days at room

586 temperature. These saturated cultures were then transferred to two 96-well microtiter plates, for

587 growth rate and stress tolerance phenotyping. These plates were incubated overnight. Cells were

588 pinned from these plates into plates for growth rate measurements. For temperature growth

589 assays, cells were pinned into four fresh YPD microtiter plates and then incubated at 0°C, 4°C,

590 10°C, and 20°C. For the microtiter plates at 0°C, 4°C, and 10°C, OD was measured at least once

26 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

591 a day for two weeks or until a majority of the strains had reached stationary phase. Growth on

592 different carbon sources was measured at 20°C in MM media with 2% of the respective carbon

593 source. Carbon sources tested were: glucose, galactose, raffinose, maltose, maltotriose, ethanol,

594 and glycerol. OD was read every two hours for one week or until saturation. All phenotyping

595 was completed in biological triplicate. The carbon source data was truncated to 125 hours to

596 remove artifacts due to evaporation. Growth curves were analyzed using the package grofit

597 (Kahm et al. 2010) in R to measure saturation and growth rate. We then averaged each strain

598 over the triplicates. We used an ANOVA corrected with Tukey’s HSD to test for growth rate

599 interactions between subpopulation and carbon source or subpopulation and temperature. We

600 used the R package pvclust (Suzuki and Shimodaira 2006) to cluster and build heatmaps of

601 growth rate by subpopulation.

602 Heat shock was completed by pelleting 200μl saturated culture, removing supernatant,

603 resuspending in 200μl YPD pre-heated to 37°C, and incubating for one hour at 37°C, with a

604 room temperature control. Freeze-thaw tolerance was tested by placing saturated YPD cultures in

605 a dry ice ethanol bath for two hours, with a control that was incubated on ice. After stress, the

606 strains were serially diluted 1:10 three times and pinned onto solid YPD. These dilution plates

607 were then photographed after 6 and 18 hours. CellProfiler (Lamprecht et al. 2007) was

608 used to calculate the colony sizes after 18 hours, and the 3rd (1:1000) dilutions were used for

609 downstream analyses. The heat shock measurements were normalized by the room temperature

610 controls, and the freeze thaw measurements were normalized by the ice incubation controls.

611 Statistical interactions of subpopulations and stress responses were tested as above.

612

613 Data Availability:

27 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

614 All short-read genome sequencing data has been deposited in the NCBI Short Read Archive

615 under the PRJNA555221. Accessions of public data is given in Table S1.

616

617 Acknowledgements:

618 We thank Francisco A. Cubillos for coordinating publication with their study; Sean D. Schoville,

619 José Paulo Sampaio, Paula Gonçalves, and members of the Hittinger Lab, in particular

620 EmilyClare P. Baker, for helpful discussion and feedback; Amanda B. Hulfachor and Martin

621 Bontrager for preparing a subset of Illumina libraries; the University of Wisconsin

622 Biotechnology Center DNA Sequencing Facility for providing Illumina sequencing facilities and

623 services; the Agricultural Research Service (ARS) NRRL collection, Christian R. Landry,

624 Ashley Kinart, and Drew T. Doering for strains included in phenotyping; Huu-Vang Nguyen for

625 strains used for hybrid genome comparison; and Leslie Shown, Anita R. & S. Todd Hittinger,

626 EmilyClare P. Baker, and Ryan V. Moriarty for collecting samples and/or isolating strains. This

627 material is based upon work supported by the National Science Foundation under Grant Nos.

628 DEB-1253634 (to CTH) and DGE-1256259 (Graduate Research Fellowship to QKL), the USDA

629 National Institute of Food and Agriculture Hatch Project No. 1003258 to CTH, and in part by the

630 DOE Great Lakes Bioenergy Research Center (DOE BER Office of Science Nos. DE-

631 SC0018409 and DE-FC02-07ER64494 to Timothy J. Donohue). QKL was also supported by the

632 Predoctoral Training Program in Genetics, funded by the National Institutes of Health

633 (5T32GM007133). DP is a Marie Sklodowska-Curie fellow of the European Union’s Horizon

634 2020 research and innovation program (Grant Agreement No. 747775). DL was supported by

635 CONICET (PIP11220130100392CO), FONCyT (PICT 3677, PICT 2542), Universidad Nacional

636 del Comahue (B199), and NSF-CONICET grant. CTH is a Pew Scholar in the Biomedical

28 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

637 Sciences and H. I. Romnes Faculty Fellow, supported by the Pew Charitable Trusts and Office of

638 the Vice Chancellor for Research and Graduate Education with funding from the Wisconsin

639 Alumni Research Foundation (WARF), respectively.

640

641 Author Contributions:

642 QKL, DL, and CTH, conceived of the study; QLK, DP, JIE, DL, and CTH refined concept and

643 design; QKL performed all population genomic and ecological niche analyses; QKL and DAO

644 sequenced genomes, performed phenotyping and statistical analyses, and mentored KVB and

645 MJ; DL and CTH supervised the study; KVB, KS, and MJ isolated and/or identified North

646 American strains; JIE, MER, CAL, and DL isolated and/or provided South American strains; and

647 QKL and CTH wrote the manuscript with editorial input from all co-authors.

648

649 Conflict of Interest Disclosure:

650 Commercial use of Saccharomyces eubayanus strains requires a license from WARF or

651 CONICET. Strains are available for academic research under a material transfer agreement.

652 653 References: 654 Almeida P, Gonçalves C, Teixeira S, Libkind D, Bontrager M, Masneuf-Pomarède I, Albertin W, 655 Durrens P, Sherman DJ, Marullo P, et al. 2014. A Gondwanan imprint on global diversity 656 and domestication of wine and cider yeast Saccharomyces uvarum. Nat Commun 5: 4044. 657 Alsammar HF, Naseeb S, Brancia LB, Gilman RT, Wang P, Delneri D. 2019. Targeted 658 metagenomics approach to capture the biodiversity of Saccharomyces genus in wild 659 environments. Environ Microbiol Rep 11: 206–214. 660 Baas-Becking L. 1934. Geobiologie; of inleiding tot de milieukunde. WP Van Stockum & Zoon 661 NV. 662 Baker E, Wang B, Bellora N, Peris D, Hulfachor AB, Koshalek JA, Adams M, Libkind D, 663 Hittinger CT. 2015. The genome sequence of Saccharomyces eubayanus and the 664 domestication of lager-brewing yeasts. Mol Biol Evol 32: 2818–2831. 665 Baker EP, Peris D, Moriarty R V, Li XC, Fay JC, Hittinger CT. 2019. Mitochondrial DNA and 666 temperature tolerance in lager yeasts. Sci Adv 5: eaav1869.

29 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

667 Bing J, Han P-J, Liu W-Q, Wang Q-M, Bai F-Y. 2014. Evidence for a Far East Asian origin of 668 lager beer yeast. Curr Biol 24: R380-1. 669 Brouwers N, Brickwedde A, Gorter de Vries AR, van den Broek M, Weening SM, van den 670 Eijnden L, Diderich JA, Bai F-Y, Pronk JT, Daran J-MG. 2019. Maltotriose consumption 671 by hybrid Saccharomyces pastorianus is heterotic and results from regulatory cross-talk 672 between parental sub-genomes. bioRxiv 679563. 673 Csardi G, Nepusz T. 2006. The igraph software package for complex network research. 674 InterJournal 1695: 1–9. 675 Czekanski-Moir JE, Rundell RJ. 2019. The Ecology of Nonecological Speciation and 676 Nonadaptive Radiations. Trends Ecol Evol 34: 400–415. 677 de Wit R, Bouvier T. 2006. “Everything is everywhere, but, the environment selects”; what did 678 Baas Becking and Beijerinck really say? Environ Microbiol 8: 755–758. 679 Dray S, Dufour A-B. 2007. The ade4 Package: Implementing the Duality Diagram for 680 Ecologists. J Stat Softw 22: 1–20. 681 Eberlein C, Hénault M, Fijarczyk A, Charron G, Bouvier M, Kohn LM, Anderson JB, Landry 682 CR. 2019. Hybridization is a recurrent evolutionary stimulus in wild yeast speciation. Nat 683 Commun 10. 684 Eizaguirre JI, Peris D, Rodríguez ME, Lopes CA, De Los Ríos P, Hittinger CT, Libkind D. 2018. 685 Phylogeography of the wild Lager-brewing ancestor (Saccharomyces eubayanus) in 686 Patagonia. Environ Microbiol 20: 3732–3743. 687 Francesca N, Canale DE, Settanni L, Moschetti G. 2012. Dissemination of wine-related yeasts by 688 migratory birds. Environ Microbiol Rep 4: 105–112. 689 Francesca N, Carvalho C, Sannino C, Guerreiro M a, Almeida PM, Settanni L, Massa B, 690 Sampaio JP, Moschetti G. 2014. Yeasts vectored by migratory birds collected in the 691 Mediterranean island of Ustica and description of Phaffomyces usticensis f.a. sp. nov., a 692 new species related to the cactus ecoclade. FEMS Yeast Res 14: 910–21. 693 Gallone B, Steensels J, Prahl T, Soriaga L, Saels V, Herrera-Malaver B, Merlevede A, 694 Roncoroni M, Voordeckers K, Miraglia L, et al. 2016. Domestication and Divergence of 695 Saccharomyces cerevisiae Beer Yeasts. Cell 166: 1397-1410.e16. 696 Gayevskiy V, Goddard MR. 2016. Saccharomyces eubayanus and Saccharomyces arboricola 697 reside in North Island native New Zealand forests. Environ Microbiol 18: 1137–1147. 698 Gibson B, Liti G. 2015. Saccharomyces pastorianus: genomic insights inspiring innovation for 699 industry. Yeast 32: 17–27. 700 Gibson BR, Storgårds E, Krogerus K, Vidgren V. 2013. Comparative physiology and 701 fermentation performance of Saaz and Frohberg lager yeast strains and the parental species 702 Saccharomyces eubayanus. Yeast 30: 255–266. 703 Gillespie RG, Baldwin BG, Waters JM, Fraser CI, Nikula R, Roderick GK. 2012. Long-distance 704 dispersal: a framework for hypothesis testing. Trends Ecol Evol 27: 47–56. 705 Gonçalves M, Pontes A, Almeida P, Barbosa R, Serra M, Libkind D, Hutzler M, Gonçalves P, 706 Sampaio JP. 2016. Distinct Domestication Trajectories in Top- Fermenting Beer Yeasts and 707 Wine Yeasts. Curr Biol 26: 1–12. 708 Hijmans RJ, Williams E, Vennes C. 2019. Package “geosphere” Type Package Title Spherical 709 Trigonometry. 710 Hittinger CT. 2013. Saccharomyces diversity and evolution: a budding model genus. Trends 711 Genet 29: 309–17. 712 Hittinger CT, Steele JL, Ryder DS. 2018. Diverse yeasts for diverse fermented beverages and

30 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

713 foods. Curr Opin Biotechnol 49: 199–206. 714 Huson DH, Bryant D. 2006. Application of Phylogenetic Networks in Evolutionary Studies. Mol 715 Biol Evol 23: 254–267. 716 Jombart T. 2008. adegenet: a R package for the multivariate analysis of genetic markers. 717 Bioinformatics 24: 1403–1405. 718 Kahm M, Hasenbrink G, Lichtenberg-Fraté H, Ludwig J, Kschischo M. 2010. grofit : Fitting 719 Biological Growth Curves with R. J Stat Softw 33: 1–21. 720 Kass JM, Vilela B, Aiello-Lammens ME, Muscarella R, Merow C, Anderson RP. 2018. 721 Wallace : A flexible platform for reproducible modeling of species niches and distributions 722 built for community expansion ed. R.B. O’Hara. Methods Ecol Evol 9: 1151–1156. 723 Kuehne HA, Murphy HA, Francis CA, Sniegowski PD. 2007. Allopatric Divergence, Secondary 724 Contact, and Genetic Isolation in Wild Yeast Populations. Curr Biol 17: 407–411. 725 Lamprecht MR, Sabatini DM, Carpenter AE. 2007. CellProfilerTM: Free, versatile software for 726 automated biological image analysis. Biotechniques 42: 71–75. 727 Lawson DJ, Hellenthal G, Myers S, Falush D. 2012. Inference of population structure using 728 dense haplotype data. PLoS Genet 8: e1002453. 729 Leducq J-B, Charron G, Samani P, Dubé AK, Sylvester K, James B, Almeida P, Sampaio JP, 730 Hittinger CT, Bell G, et al. 2014. Local climatic adaptation in a widespread microorganism. 731 Proc Biol Sci 281: 20132472. 732 Leducq J-B, Nielly-Thibault L, Charron G, Eberlein C, Verta J-P, Samani P, Sylvester K, 733 Hittinger CT, Bell G, Landry CR. 2016. Speciation driven by hybridization and 734 chromosomal plasticity in a wild yeast. Nat Microbiol 1: 1–10. 735 Libkind D, Hittinger CT, Valério E, Gonçalves C, Dover J, Johnston M, Gonçalves P, Sampaio 736 JP. 2011. Microbe domestication and the identification of the wild genetic stock of lager- 737 brewing yeast. Proc Natl Acad Sci U S A 108: 14539–44. 738 Liti G, Carter DM, Moses AM, Warringer J, Parts L, James SA, Davey RP, Roberts IN, Burt A, 739 Koufopanou V, et al. 2009. Population genomics of domestic and wild yeasts. Nature 458: 740 337–341. 741 Marsit S, Leducq JB, Durand É, Marchant A, Filteau M, Landry CR. 2017. Evolutionary biology 742 through the lens of budding yeast comparative genomics. Nat Rev Genet 18: 581–598. 743 Mathiasen P, Premoli AC. 2010. Out in the cold: Genetic variation of Nothofagus pumilio 744 (Nothofagaceae) provides evidence for latitudinally distinct evolutionary histories in austral 745 South America. Mol Ecol 19: 371–385. 746 McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, Garimella K, 747 Altshuler D, Gabriel S, Daly M, et al. 2010. The Genome Analysis Toolkit: A MapReduce 748 framework for analyzing next-generation DNA sequencing data. Genome Res 20: 1297– 749 1303. 750 Naseeb S, Alsammar H, Burgis T, Donaldson I, Knyazev N, Knight C, Delneri D. 2018. Whole 751 Genome Sequencing, de Novo Assembly and Phenotypic Profiling for the New Budding 752 Yeast Species Saccharomyces jurei. G3 8: 2967–2977. 753 Nespolo RF, Villarroel CA, Oporto CI, Tapia SM, Vega F, Urbina K, De Chiara M, Mozzachiodi 754 S, Mikhalev E, Thompson D, et al. 2019. An Out-of-Patagonia dispersal explains most of 755 the worldwide genetic distribution in Saccharomyces eubayanus. Submiss. 756 Nguyen HV, Boekhout T. 2017. Characterization of Saccharomyces uvarum (Beijerinck, 1898) 757 and related hybrids: Assessment of molecular markers that predict the parent and hybrid 758 genomes and a proposal to name yeast hybrids. FEMS Yeast Res 17: 1–19.

31 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

759 Nielsen R, Akey JM, Jakobsson M, Pritchard JK, Tishkoff S, Willerslev E. 2017. Tracing the 760 peopling of the world through genomics. Nature 541: 302–310. 761 Peris D, Langdon QK, Moriarty R V, Sylvester K, Bontrager M, Charron G, Leducq JB, Landry 762 CR, Libkind D, Hittinger CT. 2016. Complex Ancestries of Lager-Brewing Hybrids Were 763 Shaped by Standing Variation in the Wild Yeast Saccharomyces eubayanus. PLoS Genet 764 12. 765 Peris D, Sylvester K, Libkind D, Gonçalves P, Sampaio JP, Alexander WG, Hittinger CT. 2014. 766 Population structure and reticulate evolution of Saccharomyces eubayanus and its lager- 767 brewing hybrids. Mol Ecol 23: 2031–2045. 768 Peter J, De Chiara M, Friedrich A, Yue JX, Pflieger D, Bergström A, Sigwalt A, Barre B, Freel 769 K, Llored A, et al. 2018. Genome evolution across 1,011 Saccharomyces cerevisiae isolates. 770 Nature 556: 339–344. 771 Pfeifer B, Wittelsbuerger U. 2015. Package ‘PopGenome .’ 772 Premoli AC, Mathiasen P, Cristina Acosta M, Ramos VA. 2012. Phylogeographically 773 concordant chloroplast DNA divergence in sympatric Nothofagus s.s. How deep can it be? 774 New Phytol 193: 261–275. 775 Premoli AC, Mathiasen P, Kitzberger T. 2010. Southern-most Nothofagus trees enduring ice 776 ages: Genetic evidence and ecological niche retrodiction reveal high latitude (54°S) glacial 777 refugia. Palaeogeogr Palaeoclimatol Palaeoecol 298: 247–256. 778 Quiroga MP, Premoli AC. 2010. Genetic structure of Podocarpus nubigena (Podocarpaceae) 779 provides evidence of Quaternary and ancient historical events. Palaeogeogr Palaeoclimatol 780 Palaeoecol 285: 186–193. 781 Racimo F, Sankararaman S, Nielsen R, Huerta-Sánchez E. 2015. Evidence for archaic adaptive 782 introgression in humans. Nat Rev Genet 16: 359–371. 783 Raj A, Stephens M, Pritchard JK. 2014. fastSTRUCTURE: Variational Inference of Population 784 Structure in Large SNP Data Sets. Genetics 197: 573–589. 785 Rodríguez ME, Pérez-Través L, Sangorrín MP, Barrio E, Lopes CA. 2014. Saccharomyces 786 eubayanus and Saccharomyces uvarum associated with the fermentation of Araucaria 787 araucana seeds in Patagonia. FEMS Yeast Res 14: 948–965. 788 Salema-Oom M, Pinto VV, Gonçalves P, Spencer-Martins I. 2005. Maltotriose Utilization by 789 Industrial. Society 71: 5044–5049. 790 Sampaio JP. 2018. Microbe profile: Saccharomyces eubayanus, the missing link to lager beer 791 yeasts. Microbiol (United Kingdom) 164: 1069–1071. 792 Sampaio JP, Gonçalves P. 2017. Biogeography and Ecology of the Genus Saccharomyces. In 793 Yeasts in Natural Ecosystems: Ecology, pp. 131–157. 794 Schacherer J, Shapiro JA, Ruderfer DM, Kruglyak L. 2009. Comprehensive polymorphism 795 survey elucidates population structure of Saccharomyces cerevisiae. Nature 458: 342–5. 796 Sérsic AN, Cosacov A, Cocucci AA, Johnson LA, Pozner R, Avila LJ, Sites Jr. JW, Morando M. 797 2011. Emerging phylogeographical patterns of plants and terrestrial vertebrates from 798 Patagonia. Biol J Linn Soc 103: 475–494. 799 Shen X-X, Opulente DA, Kominek J, Zhou X, Steenwyk JL, Buh K V., Haase MAB, Wisecaver 800 JH, Wang M, Doering DT, et al. 2018. Tempo and Mode of Genome Evolution in the 801 Budding Yeast Subphylum. Cell 175: 1533-1545.e20. 802 Stefanini I, Dapporto L, Legras J-L, Calabretta A, Di Paola M, De Filippo C, Viola R, Capretti P, 803 Polsinelli M, Turillazzi S, et al. 2012. Role of social wasps in Saccharomyces cerevisiae 804 ecology and evolution. Proc Natl Acad Sci U S A 109: 13398–403.

32 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

805 Stewart JR, Lister AM. 2001. Cryptic northern refugia and the origins of the modern biota. 806 Trends Ecol Evol 16: 608–613. 807 Suzuki R, Shimodaira H. 2006. Pvclust: An R package for assessing the uncertainty in 808 hierarchical clustering. Bioinformatics 22: 1540–1542. 809 Sylvester K, Wang Q-M, James B, Mendez R, Hulfachor AB, Hittinger CT. 2015. Temperature 810 and host preferences drive the diversification of Saccharomyces and other yeasts: a survey 811 and the discovery of eight new yeast species. FEMS Yeast Res 15: fov002. 812 Yurkov A. 2017. Temporal and Geographic Patterns in Yeast Distribution. In Yeasts in Natural 813 Ecosystems: Ecology, pp. 101–130. 814 Zhang C, Dong S-S, Xu J-Y, He W-M, Yang T-L. 2019. PopLDdecay: a fast and effective tool 815 for linkage disequilibrium decay analysis based on variant call format files. Bioinformatics 816 35: 1786–1788. 817 818 Figure Legends: 819 Figure 1. S. eubayanus distribution and population structure. 820 S. eubayanus has a global distribution and two major populations with six subpopulations. (A) 821 Isolation locations of S. eubayanus strains included in the dataset. For visibility, circle size is not 822 scaled by the number of strains. Subpopulation abundance is shown as pie charts. The 823 Patagonian sampling sites have been collapsed to two locations for clarity. Details of sites and 824 subpopulations found in South America (B) and North America (C) with circle size scaled by the 825 number of strains. (D) Whole genome PCA of S. eubayanus strains and five hybrids with large 826 contributions from S. eubayanus. (E) PCA of just PA. (F) PCA of just PB and hybrid S. 827 eubayanus sub-genomes. Color legends in A and D apply to this and all other figures. 828 829 Figure 2. Phenotypic differences. 830 (A) Heat map of mean of maximum growth rate (change in OD/hour) (GR) on different carbon

831 sources by subpopulation. Warmer colors designate faster growth. (B) Heat map of log10 832 normalized growth at different temperatures by subpopulation. 833 834 Figure 3. Population genomic parameters.

835 (A) Network built with pairwise FST values < 0.8 between each subpopulation. FST values are 836 printed and correspond to line thickness, where lower values are thicker. Circle sizes correspond 837 to genetic diversity. (B) LD decay for each subpopulation (colors) and the species in whole 838 (black). 839

33 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

840 Figure 4. Genomic ancestries of NoAm and SoAm admixed lineages.

841 (A) For all 21 NoAm admixed strains, log2 ratio of the minimum PB-NoAm pairwise nucleotide 842 sequence divergence (dB-NoAm) and the minimum PA-NoAm pairwise nucleotide sequence

843 divergence (dA-NoAm) in 50-kbp windows. Colors and log2 < 0 or > 0 indicate that part of the 844 genome is more closely related to PA or PB, respectively. (B) Neighbor-Net phylogenetic 845 network reconstructed with the 571 SNPs that differentiate the NoAm strains. The scale bar 846 represents the number of substitutions per site. Collection location is noted. (C) Pairwise 847 nucleotide sequence divergence of the NoAm strain yHKS210 compared to strains from the PA- 848 1 and PA-2 subpopulations of PA in 50-kbp windows. (D) Pairwise nucleotide sequence 849 divergence of the NoAm strain yHKS210 compared to strains from the PB-1, PB-2, and PB-3

850 subpopulations of PB in 50-kbp windows. (E) log2 ratio of the minimum PB-SoAm pairwise 851 nucleotide sequence divergence (dB-SoAm) and the minimum PA-SoAm pairwise nucleotide

852 sequence divergence (dA-SoAm) in 50-kbp windows. Colors and log2 < 0 or > 0 indicate that 853 part of the genome is more closely related to PA or PB, respectively. (F) Pairwise nucleotide 854 sequence divergence of the SoAm strain compared to strains from the PA-1 and PA-2 855 subpopulations of PA in 50-kbp windows. (G) Pairwise nucleotide sequence divergence of the 856 SoAm strain compared to strains of the PB-1, PB-2, and PB-3 subpopulations of PB in 50-kbp 857 windows. 858 859 Figure 5. South American genomic diversity versus range, diversity by area, and isolation by 860 distance. 861 (A) Range and genomic diversity of South American sampling sites. Circle sizes correspond to 862 nucleotide diversity of all strains from that site, and pie proportions correspond to each 863 subpopulation’s contribution to � at each site. Latitudinal range of each subpopulation is shown 864 to the right. (B) Nucleotide diversity by subpopulation by sampling site, where larger and darker 865 circles indicate more diversity. “SA Sites” in gray show the diversity of all strains found in each 866 South American (SA) site. “World Sites” in darker gray show the nucleotide diversity of all 867 North American or non-South American strains, regardless of subpopulation, compared to South 868 American strains south or north of 43°S, aligned to mean latitude of all strains included in the 869 analysis. (C) Correlation of nucleotide diversity and the area or distance a subpopulation covers. 870 The y-axis shows the nucleotide diversity of each subpopulation, and circle sizes correspond to

34 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

871 the geographic sizes of the subpopulations on a log10 scale. Note that PA-1 (dark red) is as

872 diverse as PB-3 (dark blue) but encompasses a smaller area. (D) log10(pairwise nucleotide 873 diversity) correlated with distance between strains, which demonstrates isolation by distance. 874 Note that y-axes are all scaled the same but not the x-axes. Holarctic includes the S. eubayanus 875 sub-genome of two lager-brewing strains. Figure S8A shows the individual plots for the NoAm 876 lineage. Figure S8B shows the individual plot of PB-1. 877 878 Figure 6. Predicted climatic suitability of S. eubayanus. 879 Minimum training presence (light green) and 10th percentile training presence (dark green) based 880 on a model that includes all known S. eubayanus isolations, as well as a scenario of dispersal and 881 diversification out of Patagonia (inset and arrows). Black arrows signify diversification events, 882 dotted lines are diversification events where the population is not found in Patagonia, and 883 colored arrows are migration events for the lineage of matching color. Roman numerals order the 884 potential migration events. S. eubayanus has not been found in the wild in Europe, but it has 885 contributed to fermentation hybrids, such as lager yeasts. This scenario proposes that the last 886 common ancestor of PA and PB-Holarctic bifurcated into PA (red) and PB-Holarctic (blue), 887 which further radiated into PA-1 (dark red), PA-2 (light red), PB-1 (blue), PB-2 (lighter blue), 888 PB-3 (dark blue), and Holarctic (very light blue). At least four migration events are needed to 889 explain the locations where S. eubayanus has been found. I. The Holarctic subpopulation was 890 drawn from the PB-Holarctic gene pool and colonized the Holarctic ecozone. II. PB-1 colonized 891 the Pacific Rim, including New Zealand and Washington state, USA. III. An independent 892 dispersal event brought PB-1 to North Carolina, USA. IV. Outcrossing between PA-2 and PB-1 893 gave rise to a low-diversity admixed linage that has recently invaded a large swath of North 894 America. 895 896 Supplementary Figure 1. Additional visualizations of population structure. 897 (A) SplitsTree network tree built with 11994 SNPs with subpopulations circled and labeled. (B) 898 FineStructure co-ancestry plot built with 11994 SNPs. Bluer colors correspond to more genetic 899 similarity. Boxes have been added to label the subpopulations. (C) FastSTRUCTURE plot (K=6) 900 built with 150165 SNPs and showing the same six monophyletic subpopulations found with 901 other approaches. Only five NoAm strains were included in the fastSTRUCTURE analysis.

35 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

902 903 Supplementary Figure. 2. Additional phenotypic data. 904 (A) Violin plots of recovery from stress, normalized by controls. There were no significant

905 subpopulation by stress interactions. (B) Violin plots of log10 normalized mean growth rates of 906 each subpopulation at 0°C, 4°C, 10°C, and 20°C. * = p-val < 0.05 of interactions between Lager 907 and PA-2, PB-2, and PB-3 at 10°C; Lager and PA-1, PA-2, PB-1, PB-2, and PB-3 at 20°C; and 908 PB-2 and both PA-2 and NoAm at 20°C (C) Violin plots of mean growth rate on different carbon 909 sources (* = p-val < 0.05). (D) Heatmaps of significant subpopulation by temperature 910 interactions and (E) significant subpopulation by carbon source interactions. Warmer colors 911 indicate that the subpopulation-temperature or carbon source on the left hand had a faster growth 912 rate than the subpopulation-temperature or carbon source along the bottom; cooler colors 913 represent the reverse. Non-significant interactions, based on multiple test corrections, are in 914 white. More intense colors represent smaller p-values. 915 916 Supplementary Figure 3. Heterozygosity analyses. 917 (A) Summary of all SNPs vs SNPs called as heterozygous compared to the taxonomic type strain 918 for all 200 strains included in this study. The upper limit of the bar is the total SNP count. The 919 lower point corresponds to SNPs called as heterozygous. The horizontal line is 20k SNPs. (B) 920 Strain yHCT75 (CRUB 1946) is the only strain with > 20K heterozygous SNPs. (C) When the 921 heterozygous SNPs of yHCT75 were pseudo-phased (labeled), both phases clustered with PB-1. 922 923 Supplementary Figure 4. Additional population genomic statistics. 924 (A) Mean pairwise nucleotide diversity (� * 100) for each subpopulation across the genome in 925 50-kbp windows. (B) Tajima’s D across the genome in 50-kbp windows for each subpopulation.

926 (C) Mean FST in 50-kpb windows for each subpopulation compared to all subpopulations. (D)

927 Pairwise FST for each subpopulation compared to PB-1. 928

929 Supplementary Figure 5. Pairwise FST plots for NoAm and SoAm compared to all other 930 subpopulations.

931 Pairwise FST for the NoAm lineage (A) or SoAm strain (B) compared to all other subpopulations. 932

36 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

933 Supplementary Figure 6. The taxonomic type strain has a mosaic genome. 934 (A) Pairwise genetic divergence of the taxonomic type strain compared to each subpopulation. 935 (B) Comparison of pairwise genetic divergence of the taxonomic type strain compared to PA-1

936 and PB-1. (C) log2 divergence plot (as in Figure 4) showing regions introgressed from PA-1 in 937 the taxonomic type strain. 938 939 Supplementary Figure 7. Four S. eubayanus strains with S. uvarum nuclear introgressions. 940 (A) Depth of coverage plots of reads from four strains mapped to both the S. uvarum (Suva) and 941 S. eubayanus (Seub) reference genomes. (B) Zoom-in of region on Chromosome XIV where 942 these four strains have the same S. uvarum (purple) introgression into a S. eubayanus 943 background. (C) A PCA plot shows that these four strains belong to the PB-1 subpopulation of S. 944 eubayanus. (D) A PCA plot shows that the introgressed region from S. uvarum came from the 945 South American SA-B subpopulation of S. uvarum. 946 947 Supplementary Figure 8. Isolation by distance plots for NoAm and PB-1. 948 (A) Isolation by distance for all NoAm strains. The y-axis has been rescaled compared to Figure 949 5 for better visualization. (B) Isolation by distance for subpopulation PB-1. Comparisons with 950 strains from Cerro Ñielol are labeled. All comparisons of South American strains with non-South 951 American strains are on the right side. 952 953 Supplementary Figure 9. Additional Wallace climatic models. 954 (A) Model built using only South American isolation locations. (B) Model built using only non- 955 South American sites. (C) Comparison of models based on all known S. eubayanus collection 956 sites, only South American, or only non-South American sites. Where the models agree is in dark 957 green, where two models agree is in medium green, and where one model predicts suitability is 958 in light green. 959 960 Supplementary Table 1. Collection information for all strains whose genomes were sequenced or 961 analyzed in this study. 962

37 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

963 Supplementary Table 2. Average triplicate growth rates for various temperatures and carbon 964 sources. Note that this spreadsheet has multiple sheets. 965 966 Supplementary Table 3. K=6 output of FastSTRUCTURE. 967 968 Supplementary Table 4. Mantel test results. 969 970 Supplementary Table 5. Input and output for Wallace climatic modeling.

38 bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure 1 PA−1 PB−1 PB−3 NoAm B PA−2 PB−2 Holarctic SoAm −40

A C 50

50 45 −45

0 Latitude 35 Latitude −50

Latitude 30

−50 25 −120 −100 −80 −60 −55 Longitude

−100 0 100 200 −75 −70 −65 −60 Longitude Longitude

D 50 PB−2 PA−2 PA−1

PB−3 Holarctic SoAm 50 F PB−3 PB−1 E 0 PA−1 25 50 PB−1 PA−2 PB−2 PB−1 PA−1 PB−2 0 PB−2 PA−2 PA−1 −50 PB−3 0 PB−3 −25 PA−2 Holarctic PB1 Holarctic PC2 = 9.95% Hybrid PC2 = 7.81%

PC2 = 13.96% Hybrid NoAm −50 −50 −100 SoAm Holarctic −75 −150 −100 −50 0 50 −50 0 50 100 NoAm PC1 = 41.56% −150 PC1 = 19.03%

−100 0 100 200 PC1 = 40.06% bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure 2 A GR B GR PA−2 0.06 Lager NoAm 0.15 PA−1 Holarctic 0.05 PA−2 PAPB−−22 0.06 LagerLagerLager NoAmPB−3 NoAm 0.150.150.15 PA−1 HolarcticHolarcticHolarctic 0.1 PB−1 0.050.04 PB−2 PBLager−2 PAPAPB−2PA−−23−2 PBHolarctic−3 NoAmNoAmNoAm Raffinose Maltotriose Glycerol Ethanol Glucose Maltose Galactose 0.03 PB−1 PB−1 0.04 PBPB−2PB−2−2 0.10.10.1 Lager PA−1 0.05 Holarctic 0.02 20C 10C 0C 4C PBPB−3PB−3−3 Raffinose Maltotriose Glycerol Ethanol Glucose Maltose Galactose 0.03 PBPB−1PB−1−1 PAPA−1PA−1−1 0.050.050.05 20C 10C 0C 4C 0.020.01 20C 10C 0C 20C 4C 10C 0C 4C

0.01 Figure 3

PA−1 B A PB−3 0.6

0.78 0.69 0.67 0.72 0.43 PB−2 Subpop 0.51 PA−1 0.73 0.4 PB−1 0.64 ● PA−2

0.70 SoAm 2 r PB−1 0.59 0.78 PB−2 PA−2 PB−3 Hol Seub

0.2

0.64 0.80

0.0 ● 0KB 30KB 60KB 90KB 120KB 150KB 180KB 210KB 240KB 270KB NoAm SNP distance bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure 4 A yHDPN421 0.1 B 0.1 E SoAm yHKS 694 yHKB201 yHQL1481 North Wisconsin, yHKS 689 yHKB251 yHDPN422 yHKS 576 Carolina, USA 5 )

USA yHKS 212 m

yHKS 810 A yHKS 210 o yHQL1621 S yHDPN423 - yHKS 211 yHKS 812 B d

yHKB210 m 0 A o

yHDPN424 S

yHDP N422 - yHDP N424 A d (

yHDP N423yHDP N421 2 g o

yHKB201 l New −5 yHKB246 Brunswick, yHKB242 yHKB210 yHKB268 Canada South Carolina, yHKB218 USA yHKB218 I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI NoAm Genome Position yHKB242 C F

) yHKB246 1.5 1.5 m A o N

- yHKB251 B d 1.0 Subpop 1.0 Subpop m yHKB268 A PA−1 PA−1 o

N PA−2 PA−2 - A yHKS210 d

( 0.5 0.5 2 Pairwise Divergence (%) Divergence Pairwise Pairwise Divergence (%) Divergence Pairwise g o

l yHKS211

yHKS212 0.0 0.0

I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI yHKS576 Genome Position Genome Position D G yHKS689 1.5 1.5 yHKS694

Subpop Subpop yHKS810 1.0 1.0 PB−1 PB−1 PB−2 PB−2 yHKS812 PB−3 PB−3

0.5 0.5 Pairwise Divergence (%) Divergence Pairwise yHQL1481 (%) Divergence Pairwise

yHQL1621 0.0 0.0 I II III IV VIVIIVIII IX V X XI XII XIIIXIVXVXVI I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI I II III IV VI VII VIII IX V X XI XII XIII XIV XV XVI Genome Position Genome Position Genome Position bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure 5 A B C D PA−1 PA−2 PB−1 PB−2 PB−3 All Non-South American 0.20 ● PA−1 PA−2 NoAm PB−1 PB−2 PB−3 Hol All North American Subpopulation

● PA−1

● PA−2

● PB−1 −40 −40 ● ● PB−2 0.1000 0.15 ● ● ● PB−3 Nucleotide ● NoAm Diversity (%) 0.0 Population 0.1 Geographic ● ● Size 0.2 −45 −45 100000 0.0100 0.3 0.10 Latitude

0.4 200000 Latitude ● ● Nucleotide Diversity (%) Nucleotide Diversity

0.5

300000 0.0010

−50 −50 (%)) log(Nucleotide Diversity 0.05

400000

0.0001

−55 −55 0.00 ● ● p-val=0.564 p-val=0.0003 p-val=0.0001 p-val=0.0001 p-val=0.0001 p-val=0.0001 p-val=0.0001 PA−1 PA−2 PB−1 PB−2 PB−3 SoAm SA World Sites Sites Area Distance (10 Km^2) (Km) 0 10 20 0 200 400 0 500 10001500 0 5000 10000 0 100 200 0 500 1000 1500 0 5000 10000 −75 −70 −65 −60 Distance (Km) Longitude bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure 6

Predicted Climatic Suitability

1.0

0.5 60°N 0.0

40°N I

20°N IV

Latitude 0°

Ancestor 20°S

III 40°S

150°W 100°W 50°W 0° 50°E 100°E 150°E Longitude bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure S1

yHQL1621_Seub_Admix_NoAm yHQL1481_Seub_Admix_NoAm yHKS812_Seub_Admix_NoAmx yHKS810_Seub_Admix_NoAm yHKB242_Seub_Admix_NoAmi yHKB268_Seub_Admix_NoAm yHDPN423_LL2013 220_Seub_Admix_NoAm PB -1 PB -2 yHDPN421_LL2013 218_Seub_Admix_NoAm yHKB218_Seub_Admix_NoAm yHKB251_Seub_Admix_NoAm 620

yHKS211_Seub_Admix_NoAmm yHKB201_Seub_Admix_NoAm yHKB210_Seub_Admix_NoAm yHKS694_Seub_Admix_NoAm

yHKS576_Seub_Admix_NoAmd yHKB246_Seub_Admix_NoAm yHDPN422_LL2013 219_Seub_Admix_NoAm A yHKS212_Seub_Admix_NoAm yHKS210_Seub_Admix_NoAm

yHKS689_Seub_Admix_NoAmA yHDPN424_LL2013 221_Seub_Admix_NoAm yHAB578_C39 4_Seub_SoAm yHCT104_CRUB1974_Seub_PA1 yHAB76_CRUB2013_Seub_PA1 yHAB574_QC15_Seub_PA1 yHCT103_CRUB1973_Seub_PA1 yHCT101_ExCR10 7b_Seub_PA1 yHAB64_CRUB2004_Seub_PA1 yHAB569_B36 6_Seub_PA1 yHAB58_CRUB1999_Seub_PA1 yHCT96_CRUB1966_Seub_PA1 yHAB73_CRUB2011_Seub_PA1 557 yHQL775_NPCC1437_Seub_PA2 yHQL756_NPCC1432_Seub_PA2 yHQL729_NPCC1431_Seub_PA2 PB -3 Holarctic yHQL754_NPCC1435_Seub_PA2 yHQL751_NPCC1430_Seub_PA2 yHQL714_NPCC1429_Seub_PA2 yHQL727_NPCC1287_Seub_PA2 yHQL740_NPCC1282_Seub_PA2 yHQL715_NPCC1284_Seub_PA2 yHQL705_NPCC1439_Seub_PA2 yHQL708_NPCC1285_Seub_PA2 yHCT72_CRUB1943_Seub_PA2 yHAB119_CRUB2056_Seub_PA2 yHCT114_CRUB1980_Seub_PA2 yHAB79_CRUB2016_Seub_PA2 yHAB95_CRUB2032_Seub_PA2 yHAB63_CRUB2003_Seub_PA2 yHQL717_NPCC1443_Seub_PA2A yHQL709_NPCC1291_Seub_PA2 yHQL710_NPCC1441_Seub_PA2 494 yHCT90_CRUB1960_Seub_PA2P yHCT99_CRUB1969_Seub_PA2 yHCT46_CRUB1929_Seub_PA2 yHAB568_B35 6_Seub_PA2 yHAB567_B35 5_Seub_PA2 yHCT68_CRUB1939_Seub_PA2 yHAB104_CRUB2041_Seub_PA2 yHAB103_CRUB2040_Seub_PA2 yHAB101_CRUB2038_Seub_PA2 yHAB102_CRUB2039_Seub_PA2 yHAB577_B39 1_Seub_PA2 yHAB70_CRUB2008_Seub_PA2 yHQL755_NPCC1296_Seub_PA2 yHAB94_CRUB2031_Seub_PA2 yHQL730_NPCC1286_Seub_PA2 yHCT55_CRUB1948_Seub_PB3 yHCT54_CRUB1947_Seub_PB3 yHAB92_CRUB2029_Seub_PB3 yHCT112_S13 1_Seub_PB3 yHCT60_CRUB1953_Seub_PB3 yHAB83_CRUB2020_Seub_PB3 yHAB122_CRUB2059_Seub_PB3 431 yHAB81_CRUB2018_Seub_PB3 yHAB82_CRUB2019_Seub_PB3 yHAB114_CRUB2051_Seub_PB3 yHCT52_CRUB1936_Seub_PB3 yHAB491_TC50_Seub_PB3 yHCT56_CRUB1949_Seub_PB3 SoAm yHAB85_CRUB2022_Seub_PB3 yHAB84_CRUB2021_Seub_PB3 yHCT58_CRUB1951_Seub_PB3 yHCT59_CRUB1952_Seub_PB3 yHCT66_CRUB1935_Seub_PB3 yHAB118_CRUB2055_Seub_PB3 yHAB112_CRUB2049_Seub_PB3 yHCT113_S14 1_Seub_PB3 yHAB86_CRUB2023_Seub_PB3 yHAB115_CRUB2052_Seub_PB3 yHAB113_CRUB2050_Seub_PB3 yHAB120_CRUB2057_Seub_PB3 yHAB80_CRUB2017_Seub_PB3 yHAB78_CRUB2015_Seub_PB3 yHAB69_CRUB2007_Seub_PB3 367 yHCT115_CRUB1981_Seub_PB3 yHCT105_CRUB1975_Seub_PB3 yHAB565_B32 1_Seub_PB3 yHAB125_CRUB1892_Seub_PB2 yHCT100_CRUB1970_Seub_PB2 yHCT89_CRUB1959_Seub_PB2 NoAm yHAB67_CRUB2006_Seub_PB2 yHAB74_CRUB2012_Seub_PB2 yHAB66_CRUB2005_Seub_PB2 yHCT88_CRUB1958_Seub_PB2 yHCT91_CRUB1961_Seub_PB2 yHCT97_CRUB1967_Seub_PB2 yHCT117_CRUB1983_Seub_PB2 yHAB56_CRUB1997_Seub_PB2 yHAB126_CRUB2062_Seub_PB2 yHAB124_CRUB2061_Seub_PB2 yHQL733_NPCC1442_Seub_PB2 yHQL747_NPCC1301_Seub_PB2 yHQL743_NPCC1444_Seub_PB2 yHQL758_NPCC1292_Seub_PB2 yHQL744_NPCC1311_Seub_PB2 304 yHCT71_CRUB1942_Seub_PB2 yHCT70_CRUB1941_Seub_PB2 yHAB111_CRUB2048_Seub_PB2 yHAB110_CRUB2047_Seub_PB2 yHCT107_CRUB1977_Seub_PB2 yHQL736_NPCC1440_Seub_PB2 yHQL732_NPCC1294_Seub_PB2 yHQL738_NPCC1297_Seub_PB2 yHQL726_NPCC1445_Seub_PB2 yHCT119_CRUB1985_Seub_PB2 yHAB564_B31 3_Seub_PB3 yHAB566_B33 5_Seub_PB3 yHCT47_CRUB1930_Seub_PB1 yHCT116_CRUB1982_Seub_PB1 PA -2 yHAB106_CRUB2043_Seub_PB1 yHAB117_CRUB2054_Seub_PB1 yHCT57_CRUB1950_Seub_PB1 yHAB116_CRUB2053_Seub_PB1 yHCT102_CRUB1972_Seub_PB3 yHAB105_CRUB2042_Seub_PB3 yHCT84_CRUB1954_Seub_PB1 241 yHCT49_CRUB1932_Seub_PB1 yHAB91_CRUB2028_Seub_PB1 yHAB90_CRUB2027_Seub_PB1

yHCT106_CRUB1976_SuvaXSeubPB yHCT51_SuvaXSeub yHCT75_CRUB1946_Seub_PB1 yHCT64_CRUB1762_Seub_PB1 yHCT50_CRUB1933_Seub_PB1 yHAB96_CRUB2033_Seub_PB1 yHAB88_CRUB2025_SuvaXSeub yHAB87_CRUB2024_SuvaXSeub yHAB108_CRUB2045_Seub_PB1 yHAB489_TC48_Seub_PB1 yHAB109_CRUB2046_Seub_PB1 yHAB89_CRUB2026_Seub_PB1 yHAB107_CRUB2044_Seub_PB1 yHCT48_CRUB1931_Seub_PB1 yHCT76_CRUB1568T_Seub_PB1 yHCT129_1568.2_Seub_PB1 yHCT160_CRUB1568_Seub_PB1 yHAB572_CB9 2_Seub_PB1 178 yHAB573_CB9 5_Seub_PB1 yHAB571_CB8 5_Seub_PB1 yHAB570_CB7 2 yHAB99_CRUB2036_Seub_PB1 yHAB100_CRUB2037_Seub_PB1 yHCT69_CRUB1940_Seub_PB1 yHAB97_CRUB2034_Seub_PB1 yHCT63_CRUB1761_Seub_PB1 yHKB35_Seub_PB1 yHCT98_CRUB1968_Seub_PB1 yHCT87_CRUB1957_Seub_PB1 yHCT120_CRUB1986_Seub_PB1 yHAB62_CRUB2002_Seub_PB1 yHAB59_CRUB2000_Seub_PB1 yHAB77_CRUB2014_Seub_PB1 yHCT93_CRUB1963_Seub_PB1 yHCT108_CRUB1581_Seub_PB1 yHAB68_CRUB1579_Seub_PB1 yHAB57_CRUB1998_Seub_PB1 yHAB55_CRUB1996_Seub_PB1 yHCT95_CRUB1965_Seub_PB1 115 yHAB54_CRUB1995_Seub_PB1 yHCT61_CRUB1576_Seub_PB1 yHAB65_CRUB1575_Seub_PB1 yHCT92_CRUB1962_Seub_PB1 yHCT62_CRUB1593_Seub_PB1 yHAB75_CRUB1582_Seub_PB1 yHAB61_CRUB2001_Seub_PB1 yHAB505_TC64_Seub_PB1 yHCT94_CRUB1964_Seub_PB1 yHAB71_CRUB2009_Seub_PB1 yHCT53_CRUB1938_Seub_PB1 yHAB93_CRUB2030_Seub_PB1 yHAB121_CRUB2058_Seub_PB1 yHAB576_NR9_Seub_PB1 yHAB575_NR7_Seub_PB1 P1C1_Seub_PB1 yHRVM83_Seub_PB1 yHKS509_Seub_PB1 yHRVM108_Seub_Holarctic yHRVM107_Seub_Holarctic WY2042_yHQL627_BEER_lager 51.7 WLP862_yHQL618_BEER_lager PA -1 Y 12693_FM471_BEER_lager Saaz Y 1527_yHQL580_FRUIT_SuvaXSeub Y 12646_yHQL572_FRUIT_SuvaXSeub CitKP_yHQL694_FRUIT_lager Saaz Y 846_yHQL560_FRUIT_ScerXSkudXSeub CDFM21L.1_Seub_Holarctic 2 A1 A2 A2 A2 A1 P P P P P Saaz Saaz A1 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A2 A1 A1 A2 A2 A2 A2 A2 A2 A1 A2 A1 A2 A2 A2 A2 A2 A1 A2 A2 P A2 A2 A2 A2 A1 A1 P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P aXSeub v aXSeub aXSeub aXSeub aXSeub aXSeub v v v v v 1_Seub_ 5_Seub_ 6_Seub_ 6_Seub_ 5_Seub_PB3 1_Seub_PB3 3_Seub_PB3 5_Seub_PB1 5_Seub_PB1 2_Seub_PB1 Holarctic & PB -1 PB -2 PB1_Seub_PB3 -3 1_Seub_PB3 PA -2 PA -1Admix 7b_Seub_ 4_Seub_SoAm P1C1_Seub_PB1 yHAB570_CB7 VM83_Seub_PB1 UIT_lager UIT_Su UIT_Su R R R R yHKB35_Seub_PB1 yHCT51_Su yHKS509_Seub_PB1 UB2011_Seub_ UB1999_Seub_ UB2032_Seub_ UB2016_Seub_ UB2003_Seub_ UB2008_Seub_ UB2004_Seub_ UB2031_Seub_ UB2013_Seub_ UB2056_Seub_ UB2041_Seub_ UB2039_Seub_ UB2038_Seub_ UB2040_Seub_ UB1966_Seub_ UB1943_Seub_ UB1939_Seub_ UB1969_Seub_ UB1960_Seub_ UB1929_Seub_ UB2036_Seub_PB1 UB2002_Seub_PB1 UB2034_Seub_PB1 UB1998_Seub_PB1 UB1579_Seub_PB1 UB2014_Seub_PB1 UB2000_Seub_PB1 UB2027_Seub_PB1 UB2028_Seub_PB1 UB1980_Seub_ UB2020_Seub_PB3 UB1996_Seub_PB1 UB2012_Seub_PB2 UB2006_Seub_PB2 UB2019_Seub_PB3 UB2018_Seub_PB3 UB1995_Seub_PB1 UB2005_Seub_PB2 UB2033_Seub_PB1 UB2023_Seub_PB3 UB2021_Seub_PB3 UB2022_Seub_PB3 UB1582_Seub_PB1 UB1575_Seub_PB1 UB1997_Seub_PB2 UB1974_Seub_ UB2026_Seub_PB1 UB2009_Seub_PB1 UB2001_Seub_PB1 UB1973_Seub_ UB2007_Seub_PB3 UB2015_Seub_PB3 UB2017_Seub_PB3 UB2029_Seub_PB3 UB2030_Seub_PB1 UB2054_Seub_PB1 UB2043_Seub_PB1 UB2037_Seub_PB1 UB2053_Seub_PB1 UB1892_Seub_PB2 UB2042_Seub_PB3 UB2059_Seub_PB3 UB2051_Seub_PB3 UB2050_Seub_PB3 UB2052_Seub_PB3 UB2061_Seub_PB2 UB2062_Seub_PB2 UB2044_Seub_PB1 UB2046_Seub_PB1 UB2045_Seub_PB1 UB2057_Seub_PB3 UB2055_Seub_PB3 UB2048_Seub_PB2 UB2049_Seub_PB3 UB2058_Seub_PB1 UB2047_Seub_PB2 UB1930_Seub_PB1 UB1940_Seub_PB1 UB1950_Seub_PB1 UB1957_Seub_PB1 UB1968_Seub_PB1 UB1761_Seub_PB1 UB1953_Seub_PB3 UB1963_Seub_PB1 UB1959_Seub_PB2 UB1932_Seub_PB1 UB1954_Seub_PB1 UB1965_Seub_PB1 UB1762_Seub_PB1 UB1946_Seub_PB1 UB1949_Seub_PB3 UB1936_Seub_PB3 UB1576_Seub_PB1 UB1958_Seub_PB2 UB1933_Seub_PB1 UB1593_Seub_PB1 UB1962_Seub_PB1 UB1942_Seub_PB2 UB1967_Seub_PB2 UB1961_Seub_PB2 UB1931_Seub_PB1 UB1935_Seub_PB3 UB1952_Seub_PB3 UB1951_Seub_PB3 UB1964_Seub_PB1 UB1941_Seub_PB2 UB1947_Seub_PB3 UB1948_Seub_PB3 UB1938_Seub_PB1 UB1982_Seub_PB1 UB1986_Seub_PB1 UB1972_Seub_PB3 UB1581_Seub_PB1 UB1970_Seub_PB2 UB1983_Seub_PB2 UB1568_Seub_PB1 UB1975_Seub_PB3 UB1981_Seub_PB3 UB1985_Seub_PB2 UB1977_Seub_PB2 yH R R R R R R R R R R R R R R R R R R R R UB2024_Su UB2025_Su R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R R UB1976_Su R R R R R R R R R R R Hybrids UB1568T_Seub_PB1 R R R VM107_Seub_Holarctic VM108_Seub_Holarctic R UIT_ScerXSkudXSeub 221_Seub_Admix_NoAm 219_Seub_Admix_NoAm 218_Seub_Admix_NoAm 220_Seub_Admix_NoAm R R R yHAB575_NR7_Seub_PB1 yHAB576_NR9_Seub_PB1 yH yH yHAB491_TC50_Seub_PB3 yHAB489_TC48_Seub_PB1 yHAB505_TC64_Seub_PB1 yHAB574_QC15_Seub_ CDFM21L.1_Seub_Holarctic yHAB569_B36 yHAB567_B35 yHAB577_B39 yHAB568_B35 yHAB566_B33 yHAB565_B32 yHCT112_S13 yHCT113_S14 yHAB564_B31 yHCT129_1568.2_Seub_PB1 yHAB572_CB9 yHAB573_CB9 yHAB571_CB8

C yHKS689_Seub_Admix_NoAm yHKS210_Seub_Admix_NoAm yHKS212_Seub_Admix_NoAm yHKB246_Seub_Admix_NoAm yHKS576_Seub_Admix_NoAm yHKS694_Seub_Admix_NoAm yHKB210_Seub_Admix_NoAm yHKB201_Seub_Admix_NoAm yHKS211_Seub_Admix_NoAm yHKB251_Seub_Admix_NoAm yHKB218_Seub_Admix_NoAm yHKB268_Seub_Admix_NoAm yHKB242_Seub_Admix_NoAm yHKS810_Seub_Admix_NoAm yHKS812_Seub_Admix_NoAm yHAB578_C39 yHQL1481_Seub_Admix_NoAm yHQL1621_Seub_Admix_NoAm yHAB73_C yHCT96_C yHAB58_C yHCT72_C yHAB95_C yHAB79_C yHCT68_C yHAB63_C yHAB70_C yHCT99_C yHCT90_C yHAB64_C yHAB94_C yHAB76_C yHCT46_C yHCT47_C yHCT69_C yHAB99_C yHCT57_C yHAB62_C yHCT87_C yHCT98_C yHCT63_C yHAB97_C yHCT60_C yHAB57_C yHAB68_C yHCT93_C yHAB77_C yHAB59_C yHCT89_C yHAB90_C yHAB91_C yHCT49_C yHCT84_C yHAB83_C yHCT95_C yHAB55_C yHAB74_C yHAB67_C yHCT64_C yHCT75_C yHCT56_C yHCT52_C yHAB82_C yHAB81_C yHCT61_C yHAB54_C yHCT88_C yHAB66_C yHAB96_C yHCT50_C yHAB86_C yHAB84_C yHAB85_C yHAB75_C yHCT62_C yHCT92_C yHAB65_C yHCT71_C yHAB56_C yHCT97_C yHCT91_C yHCT48_C yHAB89_C yHCT66_C yHCT59_C yHCT58_C yHAB71_C yHCT94_C yHAB61_C yHCT70_C yHAB69_C yHAB78_C yHAB80_C yHAB92_C yHCT54_C yHCT55_C yHAB93_C yHCT53_C WLP862_yHQL618_BEER_lager WY2042_yHQL627_BEER_lager yHCT114_C yHAB119_C yHCT104_C yHAB104_C yHCT103_C yHAB102_C yHAB101_C yHAB103_C yHQL756_NPCC1432_Seub_ yHQL775_NPCC1437_Seub_ yHQL714_NPCC1429_Seub_ yHQL751_NPCC1430_Seub_ yHQL754_NPCC1435_Seub_ yHQL729_NPCC1431_Seub_ yHQL708_NPCC1285_Seub_ yHQL705_NPCC1439_Seub_ yHQL715_NPCC1284_Seub_ yHQL740_NPCC1282_Seub_ yHQL727_NPCC1287_Seub_ yHAB87_C yHAB88_C yHQL710_NPCC1441_Seub_ yHQL709_NPCC1291_Seub_ yHQL717_NPCC1443_Seub_ yHQL755_NPCC1296_Seub_ yHQL730_NPCC1286_Seub_ yHCT116_C yHCT120_C yHCT102_C yHAB117_C yHAB106_C yHCT100_C yHCT108_C yHAB100_C yHAB116_C yHAB125_C yHAB105_C yHAB122_C yHCT117_C yHAB114_C yHCT160_C yHCT105_C yHCT115_C yHAB113_C yHAB115_C yHCT119_C yHCT107_C yHAB124_C yHAB126_C yHAB107_C yHAB109_C yHAB108_C yHAB120_C yHAB118_C yHAB111_C yHAB112_C yHAB121_C yHAB110_C yHQL744_NPCC1311_Seub_PB2 yHQL758_NPCC1292_Seub_PB2 yHQL743_NPCC1444_Seub_PB2 yHQL747_NPCC1301_Seub_PB2 yHQL733_NPCC1442_Seub_PB2 yHQL726_NPCC1445_Seub_PB2 yHQL738_NPCC1297_Seub_PB2 yHQL732_NPCC1294_Seub_PB2 yHQL736_NPCC1440_Seub_PB2 yHCT76_C 12693_FM471_BEER_lager yHCT101_ExCR10 yHCT106_C 1527_yHQL580_F CitKP_yHQL694_F Y 12646_yHQL572_F Y Y 846_yHQL560_F Y yHDPN422_LL2013 yHDPN421_LL2013 yHDPN423_LL2013 yHDPN424_LL2013

SoAm / Holarctic PB -2 PA -2 PA -1 PB -3 PB -1 NoAm bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

PB−3:4C D PB−3:20C Figure S2 PB−3:10C 0C 4C 10C 20C PB−3:0C FT HS B PB−2:4C A PB−2:20C PB−2:10C 5 ● PB−2:0C ●●●● ● PB−1:4C 1e−01 ● ● PB−1:20C

PB−1:10C ● ●● ● ●● ● ** * PB−1:0C ● ● ● ● ● ● ● PA−2:4C ● ● subPop subPop 0 ● ● ● ● ** *** * diff ● ● ● PA−2:20C Ethanol Glycerol● ● Maltotriose● ● ● ● ● ● ● PA−1 ● ● PA−1 Ethanol ● ● Glycerol Maltotriose PA−2:10C 0.1 ● 1e−02 ● PA−2:0C ● PA−2 ● PA−2 0.0 PA−1:4C 0.020 −0.1 ● NoAm 0.020 ● NoAm PA−1:20C * ** * Subpop|Temperature ●●●● Subpop ● PA−1:10C −5 ● PB−1 ● ●SubpopPB−1 PA−1:0C ● PA−1 ● ● ● ● PA−1 NoAm:4C 0.015 ● PB−2 PB−2 ● PA−2 0.015 1e−03 ● PA−2 NoAm:20C ● ● ● PB−3 PB−3 ● NoAm Rate) log(Growth NoAm:10C ● NoAm ● Recovery Post Stress Post Recovery ● ● NoAm:0C ● HolarcticPB−1 ● ● Holarctic ● ● PB−1 0.010−10 ● Lager:4C ● ● LagerPB−2 0.010 ● ● LagerPB−2 Lager:20C Growth Rate Growth ● ● ● PB−3 Rate Growth Lager:10C ● ● ● PB−3 ● ● ● ● ● ● ● ● ● ● ● Holarctic ● ● ● Lager:0C ● 1e−04● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Holarctic 0.005 ● ● ● ● ● ● Holarctic:4C ● ● ● 0.005 ● ● ● ● ● ● ● ● Lager ● ● ● ● ● ● ● ● ● Lager Holarctic:20C −15 ● Holarctic:10C

0.000 0.000 − 1:0C PA − 1:4C PA − 2:0C PA − 2:4C PA PB − 1:0C PB − 1:4C PB − 2:0C PB − 2:4C PB − 3:0C Lager:0C Lager:4C NoAm:0C NoAm:4C − 1:10C PA − 1:20C PA − 2:10C PA − 2:20C PA PB − 1:10C PB − 1:20C PB − 2:10C PB − 2:20C PB − 3:10C PB − 3:20C Lager:10C Lager:20C NoAm:10C NoAm:20C Holarctic:0C Holarctic:4C subPopSubpop Holarctic:10C Holarctic:20C subPopSubpop Subpop|Temperature

PB−3:Raffinose PB−3:Maltotriose PB−3:Maltose E PB−3:Glycerol PB−3:Glucose PB−3:Galactose PB−3:Ethanol PB−2:Raffinose PB−2:Maltotriose Galactose Glucose Maltose Raffinose Ethanol Glycerol Maltotriose PB−2:Maltose C PB−2:Glycerol 0.08 PB−2:Glucose PB−2:Galactose * PB−2:Ethanol 0.020 PB−1:Raffinose PB−1:Maltotriose PB−1:Maltose ● Subpop Subpop PB−1:Glycerol ● PB−1:Glucose 0.06 ● ● ● PA−1 ● PA−1 PB−1:Galactose ● ● PB−1:Ethanol 0.015 PA−2:Raffinose ● PA−2 ● PA−2 PA−2:Maltotriose PA−2:Maltose diff ● NoAm ● NoAm PA−2:Glycerol ● ● ● PA−2:Glucose ● PA−2:Galactose 0.03 0.04 ● ● ● ● ● ● ● PB−1 PB−1 PA−2:Ethanol 0.00 ● ● 0.010 PA−1:Raffinose ● ● ● ● ● PB−2 ● PB−2 PA−1:Maltotriose PA−1:Maltose −0.03 Subpop|Carbon Growth Rate Growth Growth Rate Growth * ● PA−1:Glycerol ● ● ● ● PB−3 ● PB−3 ● ● PA−1:Glucose ● ● ● PA−1:Galactose ● ● ● ● 0.02 ● Holarctic● ● ● ● ● ● Holarctic PA−1:Ethanol ● ● ● ● ● ● ● ● 0.005 ● ● ● ● ● ● ● NoAm:Raffinose ● ● ● Lager ● ● Lager NoAm:Maltotriose ● ● NoAm:Maltose NoAm:Glycerol ● ● NoAm:Glucose ● ● NoAm:Galactose * NoAm:Ethanol 0.00 0.000 Lager:Raffinose Lager:Maltotriose Lager:Maltose Subpop Subpop Lager:Glycerol Lager:Glucose Lager:Galactose Lager:Ethanol Holarctic:Raffinose Holarctic:Maltotriose Holarctic:Maltose Holarctic:Glycerol Holarctic:Glucose Holarctic:Galactose − 1:Ethanol PA − 2:Ethanol PA − 1:Maltose PA − 2:Maltose PA PB − 1:Ethanol PB − 2:Ethanol PB − 3:Ethanol PB − 1:Maltose PB − 2:Maltose PB − 3:Maltose Lager:Ethanol − 1:Glucose PA − 1:Glycerol PA − 2:Glucose PA − 2:Glycerol PA Lager:Maltose PB − 1:Glucose PB − 1:Glycerol PB − 2:Glucose PB − 2:Glycerol PB − 3:Glucose PB − 3:Glycerol NoAm:Ethanol Lager:Glycerol Lager:Glucose NoAm:Maltose NoAm:Glycerol NoAm:Glucose − 1:Raffinose PA − 2:Raffinose PA PB − 1:Raffinose PB − 2:Raffinose Lager:Raffinose − 1:Galactose PA − 2:Galactose PA PB − 1:Galactose PB − 2:Galactose PB − 3:Galactose NoAm:Raffinose Lager:Galactose − 1:Maltotriose PA − 2:Maltotriose PA Holarctic:Ethanol NoAm:Galactose PB − 1:Maltotriose PB − 2:Maltotriose PB − 3:Maltotriose Holarctic:Maltose Lager:Maltotriose Holarctic:Glycerol Holarctic:Glucose NoAm:Maltotriose Holarctic:Raffinose Holarctic:Galactose Holarctic:Maltotriose Subpop|Carbon bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure S3 B A

100000

Subpop

● Holarctic

● NoAm

● PA−1

● PA−2

● PB−1 SeubPopBintroPhase PCA 100 ● PB−2 Number of SNPs

50000 ● PB−3 C PB−3 ● SoAm

pop

PB−1 PB−2 PB−2

● PB−3 0 PC2 = 14.9% Holarctic ● Holarctic ● ● yHCT75 yHCT75 PB−1 ● ● ● ● ● ● ● ●●●●●● ● ● ● ● ●● ●● ●●●●● ●●● ●●● ●●●●● ● ● ●●●●●●●●●●● ●● ●●●●●●●●●●●●●● ●● ●● ●●●●● ●● ● ●●●●●●●●●●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●● ●●●● ●●●●●● ● ●●● ●● ●●●● ●●●●●●●●●●●●● ● ●● ●● ● ● ●●●●● ●● ●●●● ● ● ● ●●●● ● 0 −50

Strain −50 0 50 100 PC1 = 20.66% bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure S4 1.00 ●●●● ● ● ●● ● ●● ● ●● ● ● ● ● ●●●● ●●●●●● ● ●● ●●● ● ●● ●●● ●●●●●● ● ●●●●●● ● ●● ●● ●●●●● ●● ● ●●●●●● ●●● 0.83 ● ●●●● ●● ●●● ● ●●●●●●● ●● ●● ● ●● ●● ● ● ●●●●● ● ● ●● ● ●●●●● ● ● ●● ● ● ● ●● ●●●● ●●●●●●● ● ●● ●● ● ● ●● ● ●● ● ●●● ● ● ● ●● ●● ●● ● ● ●● ● ●●●●●● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● PA − 1 0.75 ● ● ● ● ● ●●● ● ● ●● ● ● ●● ● ● ●●● ● ● 0.9 C ● ● ● ● ● ● PA − 1 ● A ● 0.50 ● 0.6 ● 0.25 ● ● ● ● ● ● ● ● ● ● ● ● ● 0.3 ● ●● ●● ●●● ● ● ● ●●● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ●● ● ●●●●●● ● ● ● ● 0.11 ● ● ● ●● ●● ● ●● ● ● ● ● ●●●● ●● ● ● ●●● ● ● ● ● ●● ● ●● ● ● ● 0.00 ●● ●●● ● ● ● ● ●●● ● ●●● ● ● ●●●● ● ● ●● ● ● ● ● ● ● ● ●●● ● ●●●● ● ● ● ●●●●● ●●●● ● 0.0 ● ●●●● ●●●●●●●●● ●●●● ●●●●● ● ● ●●● ●●●●● ●●● ●●●●●●● ●●● ●●●●●● ● ● ● ● ●●●●● ● ●●● ● ●●●●● ● ●●● ●●●●●● 1.00 ●●● ● ●● ●● ● ● 0.82 ● ● ● ●● ●●● ●●● ●● ●●●●●●●●● ●●●● ●● ● ●● ●● ●●●●●●●● ●● ● ●●●● ●● ● ●● ● ●●●● ●●●● ●●●●●●●● ●●●●● ● ●●●●● ●●●● ● ●●●● ●●● ●●● ●● ●●●●●● ● ●● ● ●●● ●● ●●● ●●●●●● ● ●●● ●● ●●● ●●●● ● ●●●● ● ●●●●●● ●● ●●●● ● ●●●● ●● ● ●●●●●●●●●●● ●●● ●● ●● ● ● ● ●● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● PA − 2 0.75 ● ●●● ● ● ● ● ● ● ● ● ● ● 0.9 ● ● PA − 2 0.50 0.6 ● 0.25 ● ● ● ● ● ● ● 0.3 ● ● ● ● ● ●● ● ●● ● ● ●●● ● ● ● ● ● ●● ● ●●●● ● ●● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●●● ● ● ● ●0.13 ● ●● ●● ●●● ●●●●● ●● ●●● ●● ●●●● ● ● ●● ●● ● ●●●●●●●● ●●●● ● ● ● ●● ●● ● ●● ● ●● ●● ●● ● ●● ● ● ●● ●●●● ● ●● 0.00 ● ● ● ● ●● ● ●● ● ●● ● ● ●● ●● ●●●●● ● ● ● ●●●●● ● ●● ● ● ●●●● ● ●● ● ● ●●● ● ●●● ●●● ●●● ● ● 0.0 ●●●●● ●● ● ● ●●●● ● ● ● ●●● ● ● ● ● ●● ● ● ●●● ●●●● ● 1.00 ● ● ●● ● ● ●● ● ●● ● ● ●●● ●●●●● ●●●●●● ●● ● ●●●●●● ●●● ●● ●● ● ●● ● ●●●● ● ● ● ●● ● ●●● ●● ●●●● ●●●● ●●●● ● ●● 0.71 ● ●● ●● ● ● ● ● ● ● ● ● ● ●●●● ●● ● ● ● ● ●●●● ● ● ●● ● ● ●●● ●● ● ●●●●●● PB − 1 0.75 ● ● ● ● ●● ●●●● ●●●●●● ● ● ● ●● ●●● ● ●●●●● ●●● ● ● ●● ●● ● ●● ●● ● ●●●● ● ●●● ● ● ●● ● ● ● ●●● ● ● ● ●● ●●● ● ●●● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●●●●● ● ●● ● ● ● 0.9 PB − 1 ● ● ● ● 0.50 ● ● ● 0.6 ● ● ● 0.25 ● ● ● ● ● ● 0.3 ●● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● 0.16 ● ● ● ●● ●● ● ● ● ●● ●●● ● ●● ● ●● ● ● ● ●● ●● ●●● ● ● ●● ● ● ● ●● ● ● ●● T ●●●●●●●● ● ●● ●●●●●●● ● ● ●●●●●●●●●● ● ●●●● ● ●● ●● ●●●●●●●●●●●●● ●●●●● ●●●●●●●●●●●●●●●● ● ●● ● ●● ●●● ●● ● ●● ● ●● ●● ● ●● 0.00 ● ● ● ● ●● ● ● ● ● ●● ● ● ●● ● ● ● S ● ● ●● ● ● ● ● ● ● ●●● ●●● ● ● ● ● ●● ●●●● ● ● ●● ●● ● ●●● ● ● ●● ● ● ● 0.0 F 1.00 ●

p ● ●● ● ●●●●● ●●●●●●● ●●●● ●●●●● ●●●●● ● ●●● ●● ●●●●●● ●●● ●●● ●● ●●● ● ●● ●● ●●●●●● ●● ● ●●●●● ● ●●● 0.81 ●●● ● ●●● ●●● ● ● ● ●●●● ● ●● ● ●●●●●● ●● ● ●●●●●●●●●●●● ●●● ●●●● ● ●● ● ● ● ●●●●●●●●● ●●● ●●●● ●●● ●●● ● ●●●●●● ● ● ● ●●●● ● ● ● ● ●●● ● ● ● ●● ●● ● ● ●● ●●●● ● ●●●● ● ● ● ●●● ● ● ● ● PB − 2 0.75 ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ●● ● ●

0.9 PB − 2 0.50 0.6 ● ● 0.25 ● ● 0.3 ● ●● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ●● ●●● ● ● ● ● ● ● ● ●●● ●● ●●●●●● ● ●●●● 0.08 0.00 ● ● ●● ● ●●●● ● ●●●● ● ●●● ●●●●●● ● ● ● ● ●●●●●● ●● ● ●● ●●●●● ● ● ● ● ●●●● ●● ● ● ● ● ● ●●● ●●●● ● ● ●●●● 0.0 ● ●●●●●●●●● ●●● ● ● ● ●● ●●●●●●●●●● ●● ●●● ●● ●●●●●●●● ●●●●●● ● ● ●● ●●●●●● ●● ●●●●●● ●●● ● ●●● ●●●●●●●●●●●●●●●● ● ● ●●●● ●●●●●● ● 1.00 ● ● ● ● ●● ● ●● ● ●●● ●●●●● ● ● ● ● ●●●●●● ● ●●●●● ●●●● ● ● ● ●●● ● ●●●●● ●● ●●● ● ●●●● ● ● ●●●●●●● ● ● 0.80 ● ●● ●● ● ● ● ●● ●●● ● ●●●●●● ● ● ●●● ● ●● ● ●● ● ● ● ● ● ● ● ●●●● ●● ● ●● ●●● ● ●● ●●●● ●● ● ● ●● ●●●● ● ● ●● ●● ●● ●● ● ●● ●●●● ● ●● ● ● ●●● ●●●●●● ● ●●●●●● ● ● ● ●●●●●● 0.75 ●● ●● ● ●● ● ● ●●● ●●● ● ● ●●● ● ●● ● ● ●●●● ● ●● PB − 3 ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ●

0.9 PB − 3 0.50 0.6 ● 0.25 ● ● 0.3 ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.10 ● ●● ● ●●● ●●● ● ● ● ● ●●● ● ●●● ●● ● ●● ●● ●● ●● ● ●●●● ●● ●● ● 0.00 ●● ● ● ● ●●●●●● ●● ●● ● ● ●●● ●●●●● ●●● ●● ● ●●●●● ● ●●● ● ●● ●●● ●●●●● ●●●●● ● ●●●●●●●●● ● ● ● ●● ●● ● ●● ●●●●●● ● ●●● 0.0 ●● ● ●●●●● ●●●● ●● ● ●●● ● ●●●●● ●● ●● ● ● ●●● ●●● ●●●●●●●● ●●● ●●●●●● ●●●● ● ● ● ● ● ●●●●●●●●●●● ●● ●●●●●● ● ●● 1.00 ● ● ● ● ● ● ● 0.81 ●●●●●● ●●●●●●●● ● ●●●●●●●● ●● ● ● ● ● ●● ●●●●● ● ●●●●● ● ●●●● ● ● ●●●● ●● ●●●●●●●●●●● ●●●● ●●●●●●● ●●●●● ●● ●●●● ● ● ●●● ● ●● ● ● ●●●●●● ● ●●●●●● ●● ●●●●●●● ●●●● ●●●●●●●●●●●●●●●● ● ●●●●●●●● ● ●●●●●● ●● ●●●●● ●●● ●● ●● ● ●●●●●●●● ●● ●●● ●●●●● ●● ● ●●●●●●●● ● ●● ●● ● ●● ●● ● ● ●●● ●● ●● ● ● ●● 0.75 ● ● ● ● ● ● ● ● ● Hol 0.9 ●

Hol 0.50 ● ● ● ● ● 0.6 ● ● ● 0.25 ● ● ● ● ● ● ● ● ● ● ● ● 0.20 0.3 ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ●●● ● ● ●●●● ●●●● ● ● ●● ●● ● ● ●●●● ● ●●● ●●● ●●● ● ●● ● ● ● ● ●● ● ●● ● ●●●●● ●● ● ● ●● ●●●●● ● ● ● ● ●●●●● ● ●●●●●● ● ●● ●●●● ● ● ●●●● ● ● ●●●●● ●●●●●● ●● ●●●● ●● ● ●●●● 0.00 ● ● ●● ●● ● ● ● ● ●●● ● ●● ● ●● ● ●● ● ●● ●● ● ● ●● ● 0.0 ●● ●●●●●●●● ● ●●● ● ●● ● ● ●●● ● ● ● ● ● ● ● ●● I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI Genome Position Genome Position

3 1.00 ●●● ● ● ● ● ● ● ●●● ●●●● ● ● ●●●●●● ●●● ●●●●●● ● ●● ●●● ●●● ●●●●● ●●● ●●●●●● ● ●● ●●● ●●● ●● ●●● ●● ● ● ● ●●● ●●●● ●●●●● ●● 0.78 ●● ●●●● ●● ● ● ●● ●●●● ●● ●● ● ● ●● ● ●●●●● ● ●●● ●● ●●●● ● ● ●● ● ● ●●●●● ●●●● ● ● ● ● ●● ● ● ● ●● ● ● ●●● ●● ● ● ●● ● ● ● ●● ● ●●●● ●● ● ●● ● ● ● ● PA − 1 2 ● 0.75 ● ● ● ●● ● ●●● ●●● ● ●● ●● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● D ● ● ● ● ● ● ● ● ●● ● B 1 ● ●● ● ● ● PA − 1 0.50 ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● 0 ● ● ● ●● ● ● ● ● ●● ● ●●● ● ● 0.25 ● ● ● ● ● ● ●● ● ● ● ● ● ● ● −0.91 ●● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ●● ● ●● ● ●● −1 ● ●● ● ●● ● ●● ●● ● ● ● ● ● ● ● ●● ●● ● ● ● 0.00 ●● ● ● ● ● ● ●● ●● ● ●●● ● ●●● ● ● ●● ●● ● ● ● ● ● ●● ●● ●● ● ●● ●● ●● ●● ●● ● ● ● ● ●● ● ●● ● ●● ● ● ●●●●●●● ●● ●● ● ● ● ● ● ● ● ●● ●●● ● ● ● ●● ●●● ● ● −2 ● ● ● 1.00 ● ●●●● ●● ●● ● ●●●● ●● ● ● ● ●● ●● ●● ● ●●●●●●● ●● ● ● ● 0.78 ● ● ●●●●● ● ●●●●●●●● ●●●●●●●●●●● ●● ●●●●● ●●●● ●●●●●●●● ●●●●●●● ● ●●●●●●● ● ● ●●● ● ●●● ●●● ●●●●●● ●● ●●●●●●● ●●●● ●●●● ●●●●●●●●●●● ● ●● ● ●● ● ● ●● ●● ● ●●● ● ● ● ●●●●●●● ●● ●● ● ● ● ● ● ● ●● ● ●● ● ● ● PA − 2 ● 0.75 ●● ● ● ●●● ● ●● ● ●● ● ● ● ● ●● ● ● ● ●● ● ● ● −3 ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 3 0.50 ● 2 0.25 ●

1 PA − 2 0.00 ● ● ● ● 0 ● ● ● ● ● ● ●● 1.00 ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●●● ● ● ●● ● ● ● ●● ● ● ●● ●● ●● ● ● ●●● ● ● ●●● ● ●●● ●● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● −1.10 ● ● ● ● PB − 2 ● ●● ● ●● ● ●● ●●● ●● ● ● ●● ●● ●●●● ● ● ●● ●● ● ● ● ● ●●●● ●● ● ● ● ● ● −1 ● ●● ●●●●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● 0.75 ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ●●● ● ● ● ●● ● ●● ● ● ● ● ●●●●●● ● ● ● ● ● ●● ● ●● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ● ● ● ● ●●● ●●● ● ●● ● ● ● ● ● ●● ● ●●●●●● ● ●●● ●● ●● ● ● ● ●● ● ● ●●●●● ●● ● ●● ● ●●● ● 0.51 ● ●● ● ● ● ● ●● ● ● ● ● ● ●●● ● ●● ● ● ● ● ●●● ●● ●●●● ● ●● ●●● ● ● ●●● ●●●●●●● ● ●●● ● ● ● ●●●●● ●●●●● ● ● ● ●● ●● ● ●● ● ● ●●● ● ● ●●● ●●● ●●● 0.50 ●●● ● ● ●● ● ●● ● ● ● ● ● ●● ●● ● ● ● ● ●● ● ● ● ● ●●●● −2 ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ●●● ● ● ● ● ●●●●● ● ● ●● ● ● ●● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● −3 0.25 ● ● ● ● ● ● ● 3 0.00 2 1.00 PB − 3 1 PB − 1 0.75 ● ● ● ●●● ● ● ● ● ● ●● ●●●●● ● ● T ●● ● ●●●●● ● ● ● ● ●● ● ● ●●●● ● ●● ● ●● ● ●● ● ● ● ● ● ● ●● ●● ● ● ● ● ●● ● ●● ● ● ● ●●● ●● ●●● ● ●●● ● 0.43 S ● ● ● ● ● ● ● ● ●● ● ● 0 ● 0.50 ● ●● ●● ● ● ● ●● ● ● ● ● ●● ● ●● ● ● ● ● ●●● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● F ● ●●● ● ● ●● ● ● ● ● ● ●●●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●●● ● ● ● ●●● ● ●●● ● ● ● ●● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● −1.32 ● ● ● ●● ● ● ● ●● ●●●● ● ●● ● ● ●● ● ● ● ● ● −1 ●●● ● ●● ●●● ● ●●● ● ●● ●●●●● ●●● ●● ●● ●●● ● ● ● ●● ●● ●●●● ●●● ●●● ●● ● ●●● ●● ● ●● ● ● ● ●●●●●● ● 0.25 ●● ●● ● ●● ● ● ● ● ● ● ●● ●●● ● ●●●●● ●●●●●●●●●●●● ●●●●●●● ● ●●●●● ● ● ●●● ●● ●● ● ●● ●● ●●●● ● ●●●● ● ● ● ● ● ●● ●●● ●●●●● ● ● ●●● ●● ●●●● ●●●●●●● ●●● ●●●● ● ● ●● ●● ●● ● ●● ● ● ● ●● ●●● ●●●● ● ● ●● ●● ●● ● ● ● ●● ● ● ●●●● ●●● ● ● ● ● ●● ● ● ● Tajima's D Tajima's −2 0.00 −3 1.00 ● ● ● ● ● ● ●● 3 ● ● ● ● ● ● ● ● ● ● ●● ● ● ● 0.75 ● ● ● Hol ● ● ● ● ● ● ●● ● ● ●● ● ●● ● ●● ●● ● ● ●●● ●● 0.59 ●● ● ●●●● ●●● ●● ●●● ● ●●● ●● ●● ●●● ●●●● ●●● ●● ● ● ●● ●●● ●● ●●●● ●●●●●●●●●● ● ●●●● ●● ● ●●●●● ● ● ● ● ●● ●●● ● ●●● ●● ●● ● ●● ● ● ●●●● ● ●●●● ●● ●●●● ● ● ● ● ● ●●●●● ●● ●● ●● ● ●● ● ●● ● ●●● ●● ● ● ●● ●●● ● ●● ● ●● ● ● 2 ● ● ● ● ● ● ●●● ● ●● ●● ● ●● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● 0.50 ● ● ● ● ●● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● PB − 2 ● 1 ● ● ● ● ● ● ● ● ● ● ● ●● 0.25 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● −0.10 ● ● ●● ● ●● ●●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● 0 ● ● ● ●●●● ● ● ● ● ●●● ● ●● ●● ● ● ● ●●● ●●● ● ● ● ● ● 0.00 ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ●● ● ● ● ●● ● 1.00 −1 ● ●●● ● ● ● ● ● ● ● ● ● ●●●● ● ●● ●●● ●●● ● ● ● ● ●●● ● ● ●●● ●● ●●●●● ● ●● ●●●●● ●●● ● ● ● ●● ●● ● ●●● ● ● ●●●● ● ●● ●●● ● ●● ●● ● ●●●● ● ● ● ● ●● ●●●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● NoAm ● ●● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● −2 ● ● 0.75 ● ● ● ● ● ● ● 0.64 ● ● ● ●● ● ● ● ● ● ● ●● ● ● ●●● ● ●● ● ●● ●● ●● ● ● ● ●●● ● ● ● ●● ● ●● ●●● ● ● ● −3 0.50 ●● ●● ●●● ●● ● ● ● ● ●●● ●●● ●● ●● ●●● ● ● ● ● ●●●● ●● ●● ● ●● ●● ● ●● ● ● ●● ● ● ● ● ●●● ● ●●●● ● ● ●● ● ● ● ● ● ●●●●● ● ● ● ● ● ● 3 ● 0.25 ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● 2 ●● ●● ● ● ●● ● ●● ● ● 0.00 ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ●● ● ● ●● ● ●● ●● ●● ●● ● ● ● ●● PB − 3 1.00 1 ●● ●● ●● ● ● ● ● ●● ●● ●● ●●●●●●● ●●● ●● ●●● ● ●● ●● ● ● ●●●●● ●●●●●●● ●●●● ●● ●● ●●●●●●●●● ● ● ● ● ● ●●● ● ● 0.00 ● ●●● ● ● ●●● ● ● ●●●● ● ● ● ●●●●● ●●●●●● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● SoAm ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ●● ● ● ● ●0.64 0 ● ●● ● ● ● ●● ● 0.75 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● −1 ● ● ● ● ●● ● ●●●●● ● ● ● ● ●● ● 0.50 ● ●●● ●● ●● ● ●● ●●●● ● ● ●●● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●●● ●● ●●● ● ● ●● ●●● ● ● ● ●●●●● ● ● ● ● ● ●● ● ● ● ●●● ● ● ● ● ●●● ● ● ● ● ● ●● ● ● ● ●● ●● ● ● ● ● ● ● ●● ● ● ● ● −2 ●● ● ●● ● ● 0.25 ● ● ● −3 0.00 ● I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI Genome Position Genome Position bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure S5

A B

● ● ●●●● ● ●●● ●● ● ●●● ●●●●● ●● ●●●●● ● ●●● ●● ●● ●●● ● ● ● ● ●●●● ● ●● ●●●● ● ●●● ●● ● ● ●● ●●● ●●●●● ●● ●●● ●●●●●● ●●● ● ● ● ●● 1.00 ● ●●●●●●● ● ●●●●●● ● ● ●●● ● ● ●●●● ● ● ●● ● ● ●●● ●● ● ● ●●●●●●●● ● ●● ●● ● ●●●● ●●● ● ●●●●●● ●● ● 1.00 ● ● ●●● ● ●●● ● ●●●●● ● ●● ● ●● ● ●●●●● ●● ● ●● ● ●● ● ●● ● ● ●●●● ● ● ● ●●●●●●● ●●● ●●●● ● ● ●● ●● ●●●●●● ●●●●● ●●●●●●●● ● ● ●●●● ●● ● ● ● ● ●● ● ●●● ●● ● ●● ●●●●●●●● ● ●● ● ● ● ● ● ●● ●●●● ●● ● ● ●● ● ● ● ●● ● ● ● ● ●● ● ●●● ● ●●● ●● ● ● ●● ●● ● ●● ● ● ● ● ● ●● ●● ● ● ● ● ●●●●● ●● ● ● ● ●● ● ● ●● ● ● ● ● ● ●●● ●● ●● ● ●● ● ● ● ● ●● ● ● ● ● ● ●● ●● ● ● ●● ● ● ● ●● PA − 1 ● ● ● ● ● ●● ● ● ● ●● PA − 1 0.75 ● ● ● ●● ● ● ●● ● 0.75 ● ● ●● ● ● ● ● ● ● ● ● ● ● ● 0.88 ● ● ●●● ● ● ●● ● ● 0.83 ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● 0.50 0.50 ● ● ● ●● ● ● 0.25 0.25 ● ● ● ● 0.00 0.00 ● ● 1.00 ● ●●●●●●●●● ●●●● ●●● ●●●● ● ●●●●● ●●●●●●●● ●●●●●●●●●● ●● ●●●●●● ●●● ● ● ● ● ●●●●● ●●●● ●● ●●●●● ●●●● ● ● ● ● ●●●●●● 1.00 ● ●● ●●●●●● ● ●●●●●●●● ●●● ● ● ●●●●●●●● ● ●●●●●●●●●● ●● ●●●●●● ● ●●● ● ●●●● ●●● ●●●●● ●● ●●●● ●●● ● ● ●● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ●● ● ●●●●●● ● ●● ● ● ● ●● ● ●● ●● ●●● ●●● ●●● ●● ● ●● ●●● ● ●●●●● ●●● ● ●●● ●●● ● ● ●●●● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● PA − 2 ● ● PA − 2 0.75 ● ●●● ● ●●● ● ●● ● ● ● ●● 0.75 ● ● ● ● ● ●● ● ●● ●●● ● ●● ● ● ● ● ●● ● ●●● ● ●●● ● ● ●●● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ●●● ●● ● ● ● ● ● ● ● ● ● ● ●● 0.80 ● ● ● ● ● ● ● ● ●●●● ● ● ● ● ● ● ● ● ● 0.70 0.50 ● ● 0.50 ● ● ● ●● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ●● ● ●●● ●●● ● ● ● ●●●● ● ● ●● ●● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.25 0.25 ● ● ● ● 0.00 0.00 ●

1.00 ● ● ● ● ● ● ●● 1.00 ● ● ● ● ●●●● ●●●●● ●●●●●●●● ●●●● ●● ● ●●●● ●●● ●● ●● ●● ●●● ● ● ●● ●●●●●●●●● ●● ●●●● ●●●●●●● ●●●●●● ●●●●●●● ●●●●●●●● ●●●● ●● ●● ● ●● ●●● ●●●● ●●●●● ●●●●●●● ●●●●●●●● ●●●● ●● ●●● ●●●●●●●●●●● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ●●●● ● ● ● ●● ● ● ● ● ● ● ● PB − 1 ● ● ● ● ● ● ● ● ●● PB − 1 ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● 0.75 ● ● ● ● ● ● 0.75 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ●●● ● ●● ●● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● 0.50 ●● ●● ●● ●● ● ● ● ● ● ●●● ●● ● ● ● ●● ●● ● ● ●●● ● ● ● ●●●●● ● ● ● 0.64 0.50 ● ● ● ● ● ● ● ● ● ●● ●●● ● ●●● ● ● ●● ● ●● ● ● ● ● ● ● ●● ● ●●●● ●● ● ●● ● ●● ● ● ●● ● ●●●● ●●●●● ● ●●● ● ●●● ●● ● ●● ● ● ●● ● ● ● ●● ● ● ● ●● ●●● ●● 0.64 ● ● ● ● ● ● ● ● ● ● ● ●●● ● ●● ●●● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ●● ● ● 0.25 ● 0.25 ● ● ● ● ● ● 0.00 0.00 ● ● ●● ● ● ● ● ●●● ● ●● ● ● ● ● ●● 1.00 ● ●● ●●● ● ●●●●●● ●●●●●●●● ●●●● ●●●● ●● ● ●● ●● ●●● ● ●●● ●●●●●●● ●●●● ● ● ●●●● ●●● ●● ●●●●●●● ● ●●●●●●● ●● ●●●●●●● ●●●● 1.00 ● ● ●●●●● ● ●●●●● ●●● ●●●●●●●● ●●●● ●●● ● ● ●●●●● ●●● ●●● ● ●●●● ●● ●● ●●●● ●●● ●●●●● ●●●● ●●● ●●●●●●● ● ●●●● ● ●●●●●● ●● ●● ●●●●●●●●●●●● ●●● ●● ● ●●●● ●●● ● ● ● ●● ● ● ● ● ●● ●●● ● ●●●●● ●● ●● ● ●● ●● ● ● ●● ●● ● ●●●● ● ●●●● ●● ● ● ● ● ●●● ● ●● ● ●●● ● ●● ●●●●●●● ● ● ●●● ● ●●●● ●● ● ●●●●●● ●● ●●●●●● ●●●●● ● ●●●●● ●●● ● ● ● ● ● ● ●● ● ● ●●●● ● ●●● ● ● ●● ●● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ● ● ● ●● ● ● ● ●●● ● ● ● ● ●● ●● ● ● ● ● ● PB − 2 ● ● ● ● ● ● ● ● ● PB − 2 ●● ● ● ● ●●● ● ●●● ● ● ● ● ● ● ●● ● ● ● ● ● 0.75 ● ● ● ● ● ● ●●● ● 0.88 0.75 ● ● ● ● ● ●● 0.88 T ● ● ● ● T ● ● ● ● ● ● ●● S 0.50 S 0.50 ● ● F 0.25 F 0.25 ● 0.00 0.00 ● ● ● ● ● ●●● ● ●●●● ●●● ● ● ● ●● ● ●●●● ●●● ● ●●●●●● ● ● ● 1.00 ●●● ● ●●●● ●● ●●●●●●● ●●●●● ●●●● ● ● ●●●● ●●● ● ●● ●●●● ●●●●●● ● ● ●●● ●●●● ●●●●●● ●●●●● ● ●●●●●● ● ●● ●● 1.00 ●●●●●●● ● ● ● ● ●●●●●●●● ●●● ● ●● ●●● ● ●●●● ●● ● ●● ● ● ●●●●●●●●●● ● ●●●●● ●●●● ●● ●●●● ●●●●● ●●●● ● ● ● ●●●● ●● ●●●●● ● ● ● ● ● ●● ● ●● ● ●● ●●●●●● ● ● ●● ● ● ● ●● ●● ● ●● ●●●●● ●● ● ●● ● ● ●● ● ● ●●●●● ●●●●● ● ●●● ● ● ●●● ●●●●●● ● ● ● ● ●●●●●●●● ●● ● ●● ●●● ●● ● ● ●●● ●● ●● ● ● ● ● ● ● ● ● ●● ●● ● ●●●● ●● ● ● ●● ●● ●● ● ● ● ● ●● ●● ● ● ●● ● ● ● ● ● ● ● ● PB − 3 ● ● ● ● PB − 3 ● ● ● ● ● ●● ● ● ●● ● ●● ●● ● ● ● ●● ● ● ● ● ● ● ●● ●●● ●● ● 0.75 ● ● ● ● ● ●● ● ● 0.75 ● ● ●● ● ● ● ● ● ● 0.86 ●●● ● ● ●● ● ●●● ● ● ● ●● 0.85 ● ● ●● ● ● ●● ●●● ● ● ●● ●● ●●● ● ● ●● ● ● ● ● ●● ● ● ●● 0.50 ● ● ● ● 0.50 ● ● ● ● 0.25 0.25 ● 0.00 0.00 ● 1.00 ● ● ●● ●● ● ● ● ● 1.00 ● ● ● ● ● ● ● ● ● ●●●●●●● ●●●●● ●●●●●●●● ●●●●● ●●●● ●●●● ●●●●● ● ●●● ●●● ●●●●●●● ●●●● ●●● ●●●●● ● ●●●●●● ● ●●●●●●●●●●●●●●●●●●●● ●●●●●●●● ● ●●●●●●● ●●●●●●● ●●●●●●●● ●● ●●●● ●● ● ●●●●●● ●●● ● ●●●● ●● ● ●●●●●●●●●●● ●●●●●●●●●●●●●●● ●●●●●●●● ● ●●●● ●●●●●●●● ●● ●● ●●●●●●●●●●●●● ●● ●● ● ● ● ●● ● ● ● ●●● ●●●● ●● ●●● ● ● ● ●● ●●●● ●● ● ● ● ●●●●● ●● ● ●● ●●● ● ● ●●●● ● ● ●●●● ● ● ● ●● ● ● ●● ● ● ●●● ● ● ●●●● ●● ● ●●● ●●●● ● ● ● ●● ●●●●● ●●●●● ● ●●●● ●●●● ● ●●● ●●● ● ●●● ● ●●● ● ● ●● ●●●● ● ●●●● ● ●● ●● ● ●●●●●● ●● ● ●●● ● ●●●●●● ●● ●● ●● ● ●●●● ●● ●● ●●● ●● ●●●●●● ● ●● ●●●●●● ● ●●● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ●● 0.75 Hol 0.75 ● ● Hol ● 0.86 ● ● ● 0.86 ● ● ● 0.50 ● 0.50 ● 0.25 0.25 ● 0.00 0.00 ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●● 1.00 ● ● ● ● ● ● ● 1.00 ● ● ● ● ● ● ● ● ● NoAm ● 0.99 SoAm ● 0.99 0.75 ● 0.75 ● 0.50 0.50 0.25 0.25 0.00 ● 0.00 ● I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI Genome Position Genome Position bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure S6 A 3 B 2 PA − 1

1 1.5

0 3

1.0 2 PA − 2

0 0.5 3 (%) Divergence Pairwise Avg

2 PB − 1

1 0.0

0 I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI Genome Position 3 C 2 PB − 2

Avg Pairwise Divergence (%) Divergence Pairwise Avg 1

0 5 3

2 PB − 3

1 0

3 Divergence log2(Pairwise

Hol −5

0 I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI Genome Position Genome Position bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

SeubPopBintroPhase PCA Figure S7 100 yHCT51 C 15 A Species PB−3 10 Seub 5 Suva

Average Depth Average 0 Suva Seub Genome Position 50

yHCT106 pop

60 Species PB−1 40 PB−2 Seub PB−2 20 Suva PB−3

Average Depth Average 0 0

Suva Seub PC2 = 14.9% Holarctic Genome Position yHAB87 Holarctic

300 Species yHCT106 yHAB87 200 PB−1 Seub yHCT51 yHAB88 100 Suva −50

Average Depth Average 0 Suva Seub Genome Position yHAB88 −50 0 50 100 Species PC1 = 20.66% 40 Seub 20 SuvaSAintro PCA Suva

Average Depth Average 0 D Suva Seub yHAB87 Genome Position 30

yHCT106 yHAB87 yHAB88 B 300 200 yHCT51

100

SA−B 0 20 0 200000 400000 600000

yHAB88 50 40 pop 30 Australasia 20 HOL 10 Hybrid 0 Species 10 0 200000 400000 600000 SA−A Seub PC2 = 21.81% yHCT106 Suva SA−A SA−B 80 Mean Depth

20 0 0 0 200000 400000 600000

yHCT51

HOL 15

10 Australasia 5 −10 0 0 200000 400000 600000 −60 −40 −20 0 ChrXIV PC1 = 52.77% bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure S8

B A 0.5 0.003 Comparisons with Cerro Ñielol strains

0.4

0.002 0.3

0.2

Nucleotide Diversity (%) Nucleotide Diversity 0.001 (%) Nucleotide Diversity Comparisons of 0.1 SA strains with NZ, WA, and NC strains 0.0

0 500 1000 1500 0 5000 10000 Distance (KM) Distance (KM) bioRxiv preprint doi: https://doi.org/10.1101/709535; this version posted July 19, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

Figure S9

A B Predicted Climatic Suitability with Only South American Sites Predicted Climatic Suitability without South American Sites

1.0 1.0 60°N 60°N

0.8 0.8 40°N 40°N

20°N 0.6 20°N 0.6

Latitude 0° Latitude 0° 0.4 0.4

20°S 20°S 0.2 0.2

40°S 40°S

0.0 0.0

150°W 100°W 50°W 0° 50°E 100°E 150°E 150°W 100°W 50°W 0° 50°E 100°E 150°E Longitude Longitude

C Predicted Climatic Suitability Comparision of Three Different Strain Sets

3.0 60°N

2.5 40°N

2.0 20°N

1.5

Latitude 0°

1.0 20°S

0.5 40°S

0.0

150°W 100°W 50°W 0° 50°E 100°E 150°E Longitude