Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1935
Divergence, selection, demographic history and conservation genomics of sibling bird species in boreal forest in Northern Eurasia and the Qinghai- Tibetan Plateau
KAI SONG
ACTA UNIVERSITATIS UPSALIENSIS ISSN 1651-6214 ISBN 978-91-513-0952-1 UPPSALA urn:nbn:se:uu:diva-408978 2020 Dissertation presented at Uppsala University to be publicly examined in EBC lecture room, Norbyv 18D, Uppsala, Friday, 12 June 2020 at 13:00 for the degree of Doctor of Philosophy. The examination will be conducted in English. Faculty examiner: Professor Juha Merilä (Ecological Genetics Research Unit, University of Helsinki).
Abstract Song, K. 2020. Divergence, selection, demographic history and conservation genomics of sibling bird species in boreal forest in Northern Eurasia and the Qinghai-Tibetan Plateau. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1935. 47 pp. Uppsala: Uppsala University. ISBN 978-91-513-0952-1.
I used two pairs of sibling boreal forest bird species to study divergence, selection, demographics, and conservation in northern Eurasia and the Qinghai-Tibetan Plateau at the microsatellite level (chapter 1) and whole genome level (chapters 2, 3, 4 and 5). In chapter 1, which is the first study to describe genetic diversity of the Sichuan Jay, I used microsatellite markers to estimate genetic differentiation in Sichuan Jay and Siberian Jay populations. The results showed similar levels of genetic variability, strong population structure, and high genetic differentiation between the two species and among different populations. In chapter 2, I used demographic analyses, and found that the Chinese Grouse has experienced substantial changes in population size from the beginning of the last interglacial, with a peak just before the last glacial maximum. The results inferred from the whole genome sequencing and species distribution models support a history of population size fluctuations. In chapter 3 to 5, I used population genomic methods to explore genomic variation, demographic divergence, local adaptation, and inbreeding from 29 whole genome re-sequenced individuals of Chinese Grouse and Hazel Grouse. I found strong evidence for population structure, changing demographic histories, and varying inbreeding levels and genetic load within both species. In Chinese Grouse, an isolated population in the northern part of the species range showed the lowest genetic diversity, high pairwise FST, high LD decay, higher inbreeding and genetic load compared to two other populations. In Hazel Grouse, there were strong population differences and inbreeding levels among the three populations, especially among the Swedish and German populations. The Swedish population likely lost genetic diversity during the re-colonization of the boreal forests in Scandinavia after the last glaciation. Analyses of genetic load showed that purifying selection of mildly deleterious mutations has been more efficient in Hazel Grouse, a species with a larger population size and range compared to Chinese Grouse. However, when I compared the genetic load as the ratio between highly deleterious loss-of-function mutations and synonymous mutations for Chinese Grouse and Hazel Grouse, purifying selection did not seem to have a large effect. My findings show that small, isolated and fragmented populations of forests birds loose genetic variation and may thereby become vulnerable to future challenges and also that populations may track past habitat changes and adapt to local conditions.
Keywords: Sichuan Jay, Siberian Jay, Chinese Grouse, Hazel Grouse, genomics, demographic history, ROH, genetic load
Kai Song, Department of Ecology and Genetics, Animal ecology, Norbyvägen 18 D, Uppsala University, SE-752 36 Uppsala, Sweden.
© Kai Song 2020
ISSN 1651-6214 ISBN 978-91-513-0952-1 urn:nbn:se:uu:diva-408978 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-408978)
观书有感 半亩方塘一鉴开， 天光云影共徘徊。 问渠那得清如许， 为有源头活水来。 There lies a glassy oblong pool, Where light and shade pursue their course. How can it be so clear and cool? For water fresh comes from its source.
List of Papers
This thesis is based on the following papers, which are referred to in the text by their Roman numerals.
I Song, K., Halvarsson, P., Fang, Y., Barnaby, J., Germogenov, N., Sun, Y. H. & Höglund, J. (2020). Genetic differentiation in Si- chuan jay (Perisoreus internigrans) and its sibling species Sibe- rian jay (P. infaustus). Conservation Genetics 21: 319-327 doi:10.1007/s10592-020-01252-y. II Song, K., Gao, B., Halvarsson, P., Fang, Y., Jiang, Y. X., Sun, Y. H., & Höglund, J. Genomic analysis of demographic history and ecological niche modeling in the endangered Chinese Grouse Tetrastes sewerzowi. Submitted manuscript. III Song, K., Gao, B., Halvarsson, P., Fang, Y., Klaus, S., Jiang, Y. X., Swenson, J. E., Sun, Y. H., & Höglund, J. Conservation ge- nomics in the boreal forest: Population structure and local adap- tation in the sibling species Chinese Grouse and Hazel Grouse. Manuscript. IV Song, K., Gao, B., Halvarsson, P., Fang, Y., Klaus, S., Jiang, Y. X., Swenson, J. E., Sun, Y. H., & Höglund, J. Demographics and divergence of the sibling species Chinese Grouse and Hazel Grouse inferred from whole genome re-sequences. Manuscript. V Song, K., van der Valk, T., Gao, B., Halvarsson, P., Fang, Y., Klaus, S., Jiang, Y. X., Swenson, J. E., Sun, Y. H., & Höglund, J. Purifying selection in Grouse is more efficient in large popu- lations. Manuscript.
Reprints were made with permission from the respective publishers.
Contents
Introduction ...... 11 1.1. Glaciation, uplift of the Qinghai-Tibetan Plateau, and speciation .... 11 1.2. Conservation genomics in boreal forests ...... 13 1.3. Study species ...... 14 1.4. Objectives ...... 15 Methods ...... 16 2.1. Sample collection ...... 16 2.2. Microsatellite allele analysis ...... 17 2.3. Species distribution model (SDM) ...... 17 2.4. Population genomic analysis ...... 18 Results and discussion ...... 21 3.1. Genetic differentiation in Sichuan jay and Siberian Jay (Chapter 1) ...... 21 3.2. Genomic analysis of demographic history and ecological niche modeling in the endangered Chinese Grouse (Chapter 2) ...... 23 3.3. Conservation genomics in the boreal forest: Population structure and local adaptation in Chinese Grouse and Hazel Grouse (Chapter 3) ...... 25 3.4. Demographics and divergence of Chinese Grouse and Hazel Grouse (Chapter 4) ...... 28 3.5. Purifying selection in Grouse is more efficient in large populations (Chapter 5) ...... 30 Conclusions and future perspective ...... 32 Svensk sammanfattning ...... 35 中文结论 ...... 37 Acknowledgments...... 39 Bibliography ...... 41
Abbreviations
QTP Qinghai-Tibetan Plateau PSMC Pairwise Sequentially Markovian Coalescent SDM Species distribution model SNP Single-nucleotide polymorphism WGS Whole genome sequence AR Allelic richness He Expected heterozygosity Ho Observed heterozygosity Ne Effective population size LD Linkage disequilibrium PCA Principal component analysis ROH Runs of homozygosity MSMC Multiple Sequentially Markovian Coalescent LIG Last interglacial LGM Last glacial maximum
Introduction
1.1. Glaciation, uplift of the Qinghai-Tibetan Plateau, and speciation During the Quaternary, vast areas of the northern parts of the Boreal region were repeatedly affected by major glaciations. Ice sheets that formed over Scandinavia spread eastwards across the Northwest Russian Plains and south- wards across northern and middle Europe. The glaciations have had a signifi- cant influence on the present genetic structure of populations, species, and communities both in in Northern Eurasia and North America (Hewitt 2000; Hewitt 2004; Soltis et al. 2006; Svendsen et al. 2004). Before the Quaternary Period, the global average temperature in the mid-Pliocene (3.3–3 Mya) was 2–3º C higher than today, whereas carbon dioxide levels were the same (Robinson et al. 2008). The change to a cooler, dry, seasonal climate had con- siderable impacts on the Pliocene vegetation, reducing distribution of tropical species worldwide, while coniferous forests and tundra covered much of the north (Comins and Kaufmann 2011). The rise of the Qinghai-Tibetan Plateau (QTP) and increase in global ice volume had a great influence on the interior Asian aridification, especially for the formation of the Loess Plateau, with loess deposits, sediment formed by accumulation of windblown dust, since 2.40 Ma (An 2000; Ding et al. 1999; Li et al. 2015; Sun and An 2005). During the Late Miocene and the Pliocene periods, the QTP had an expanding forest cover (up to between 1000 meter and 2000 meter a.s.l.) (Huang and Ji 1979; Li and Fang 1999; Mulch and Chamberlain 2006; Zheng et al. 2000), particularly at its eastern edge (mostly in the Chinese provinces of Qinghai, Sichuan, and Yunnan). In the early Qua- ternary, the warm and moist forest ecological environment found in the Ter- tiary continued expanding on the Loess Plateau. During the middle and late Pleistocene, the climate gradually became drier and the vegetation changed to arid grassland. These patterns are not only consistent with the changing char- acteristics of the fossil assemblage, but also with the formation of the loess area (Ding et al. 1999; Liu 1985; Liu 1996). During the Quaternary period, the boreal forest experienced many climate- induced fluctuations (Gauthier et al. 2015). Several glacial and interglacial periods, along with the uplift of the QTP and accumulation of the Loess Plat- eau, promoted a geographic distribution pattern of boreal forest as seen today.
11 Divergence and speciation of boreal species were promoted by natural fluctu- ations in climate and boreal forest distribution in QTP, Northern Eurasia and North America (Baba et al. 2002; Hewitt 1996; Weir and Schluter 2004; Zhan et al. 2011). The historical glacial-interglacial cycles caused dramatic popula- tion size fluctuations and thus long-term declines in effective population sizes for some species, and even extinction (Hung et al. 2014; Kozma et al. 2018; Mays et al. 2018; Zhao et al. 2013). Although there is controversy about the role of the Quaternary ice ages in vertebrate speciation across the globe, the geographical distributions of some boreal species have indeed been fragmented periodically by habitat changes accompanying global cooling. Many sibling species inhabiting QTP and Northern Eurasia show adaptions to cold environments, for example boreal bird species with feathered legs and nostrils (Persons et al. 2016). The origin of the cold-adapted Pleistocene species, especially the megafauna, has usually been sought either in arctic tundra or in the cool steppes elsewhere (Kahlke 1999; Kahlke 2014). An alternative scenario called the “out of Tibet hypoth- esis”, where cold adapted species evolved on the Tibetan high plateau, has been suggested based on new fossil assemblages (Deng et al. 2011; Wang et al. 2016; Wang et al. 2014). However, few studies have been conducted on sibling species in QTP and eastern and western Siberia, especially for species that have poor dispersal capability (Lu et al. 2012; Sun et al. 2001). During the Pleistocene, many avian species living in boreal forest evolved into super- species or diverged in to different subspecies after they experienced allopatric divergence and local adaptation (Davis and Shaw 2001; Hewitt 2004; Li et al. 2013; Ren et al. 2017). As a result of the QTP uplift during the Late Miocene and the Pliocene periods, the high altitude boreal forests were preserved in the southeastern edge and the connection to the boreal forest in Siberia was lost. In present time, the QTP is one of the key high altitude biodiversity hotspots in the world (Myers et al. 2000). Although being a biodiversity hotspot, the evolutionary history and genetic diversity of sibling species in the QTP and Siberian boreal forests have not yet been fully examined.
12 1.2. Conservation genomics in boreal forests Conservation genomics is a new field in conservation biology that applies novel whole genome-scale analyses to address conservation issues, aiming to improve capacity for resource managers to protect species (Supple and Shapiro 2018). Genomic approaches are providing important insights into bi- otic responses to environmental change and has lately become more widely adopted in conservation (Colella et al. 2019; Primmer 2009). However, cli- mate fluctuations due to glaciations has had large effects on the extent of the boreal forests on a continental scale across the Eurasian subarctic and QTP. Between these regions, comparisons of genetic diversity, local adaptation, population structure and introgression, remain largely unknown (Colella et al. 2019; Fedorov et al. 2020). Conservation genomic studies on Arctic species have been limited to studies on species identity, degree of hybridization, ge- netic diversity, demographic history and effective population size (Colella et al. 2019; Ouborg et al. 2010; Supple and Shapiro 2018). Inbreeding (mating between relatives) is a major concern for conservation as it decreases individual fitness and can increase the risk of population ex- tinction by negatively affecting fitness-related characters in many species of plants and animals, including humans (Charlesworth and Willis 2009; Frankham 1995b; Kardos et al. 2018; Newman and Pilson 1997; Soulé 1980). Inbreeding may become prevalent, especially in small and isolated popula- tions (Höglund 2009). Habitat fragmentation leads to small, fragmented pop- ulations which are more likely to be inbred, which is likely to increase the chance for extinction events over time (Frankham 1995a; Frankham 1995b). Inbreeding depression is the decline in the value of a fitness trait as a direct consequence of inbreeding (Shields 1987; Wright 1977). Extinction rate due to inbreeding has been shown to be greater under harsher and more variable environmental conditions than under benign conditions (Frankham 1998). However, inbreeding becomes of particular concern to conservation biology when there are negative fitness effects on individual phenotypes, which will affect homozygosity on a genome-wide level (Höglund 2009). Reduced fit- ness by inbreeding is a dynamic property influenced by a number of evolu- tionary factors in the past and present which is only expressed in homozygotes (Hedrick and Garcia-Dorado 2016). Purging can reduce the actual fitness de- pression through increased purifying selection facilitated by inbreeding (Hedrick and Garcia-Dorado 2016).
13 1.3. Study species In this thesis, I used two pairs of sibling species inhabiting boreal forests in the QTP and Northern Eurasia, respectively, to explore their genetic differen- tiation, adaption to their local environment and demographic history. In chapter 1 of this thesis, I focus on the sibling species of Sichuan Jay (Perisoreus internigrans) and Siberian Jay (P. infaustus). The Sichuan Jay is one of the least known endemic bird species in western China and inhabits high mountain coniferous forests at an altitude between 2800 m to 4500 m. It is a sibling species to the Siberian Jay, which inhabits boreal forest from East- ern Siberia to Fennoscandia in the west. The Sichuan Jay has only been rec- orded in 17 locations in Gansu, Sichuan, Qinghai and Tibet (Jing 2005; Sun et al. 2001). The species is classified as “Vulnerable” in the IUCN red list as it has small, declining and severely fragmented populations as a result of ex- tensive deforestation throughout its range (IUCN 2020). Despite the small and declining populations, it has not been put on the national protection list in China. In chapters 2 – 5, I utilize another pair of sibling species: The Chinese Grouse (Tetrastes sewerzowi) and Hazel Grouse (T. bonasia) to study demo- graphic history, divergence history, and population genomics. The Hazel Grouse is one of the smallest species of the grouse family, and is distributed over northern and central Eurasia (Vaurie 1965). The Hazel Grouse is classi- fied as Lower Risk (Least concern) overall by the IUCN (IUCN 2020). How- ever, it is on the red-list in some central and southern European countries be- cause local populations have become extinct or declining over a large portion of its range in Europe (Rózsa et al. 2016; Swenson 1991). The sibling species, Chinese Grouse is endemic and restricted to narrow and fragmented conifer forests on the southeastern edge of the QTP (Sun 2000; Sun et al. 2003). The Chinese Grouse is a globally rare and endangered species, which is listed in Category I of Nationally Protected Animals in China as well as listed in Near Threatened with decreasing trend by the IUCN (IUCN 2020; Storch 2000; Sun et al. 2003). Many remaining populations of these two sibling species are scat- tered and small (Lu and Sun 2011; Storch 2000).
14 1.4. Objectives The aim of this thesis was: 1) using genetic and genomic data from two pairs of sibling species to explore the demographic divergence history and local adaptation of the QTP and Northern Eurasia and 2) study conservation ge- nomics of boreal forest species on a continental scale across the Eurasia sub- arctic and the QTP.
15 Methods
2.1. Sample collection In chapter 1, 58 Sichuan jay samples were obtained during 1999 to 2003 from China and 205 Siberian jay samples were collected during 2008 to 2010 from Sweden and Russia. Siberian jay blood samples came from 3 populations in Russian Siberia (i.e. Burenski; Kolyma and Yakutsk) and from 2 populations in Sweden (i.e. Ånnsjön and Vilhelmina).
Figure 1. Distribution and sampling localities of the Chinese Grouse (red) and the Hazel Grouse (blue). The fine scale distribution of Chinese Grouse (green) is inset in the lower left corner.
In chapter 2 – 5, a blood sample was collected from one Chinese grouse indi- vidual from the Lianhuashan Nature Reserve, Gansu province, China, for de novo sequencing. Then I collected 29 samples from the two sibling species from 8 locations (Figure 1). The Chinese Grouse samples (n=16) were from three populations (3 from the Qilian Mountains (QLS); 3 from Zhuoni County (ZN); and 10 from the Lihanhuashan National Nature Reserve (LHS)). The Hazel Grouse (n=13) were from five populations (1 from Northeast Poland
16 (NEP); 1 from the Austrian Alps and 3 from the Bavarian forest, Germany (GER); 3 from Jämtland, Sweden (SWE); and 5 from Northeast China (XLJ)). We also sequenced 3 ptarmigan (genus Lagopus) samples and used them as an outgroup.
2.2. Microsatellite allele analysis In chapter 1, I used microsatellite allele frequencies to report genetic variation and genetic diversity among endangered Sichuan Jay population and Siberian Jay populations. Twenty one microsatellite loci originally developed for Sibe- rian jay were amplified (Jaari et al. 2008) in our Sichuan jay and Siberian jay samples. After checking if loci were located on the Z-chromosome, presence of null alleles, Hardy and Weinberg equilibrium and linkage disequilibria, we used 11 microsatellite loci to analyze Sichuan jay samples and 9 of the same microsatellite markers out of the previous 11 to analyze the Siberian jay sam- ples. Then, we calculated expected heterozygosity (He), observed heterozygosity (Ho), FIS, allelic richness (AR), and effective population size (Ne) in each of the sampled populations. The neutral genetic diversity for each locality was calculated in FSTAT 2.9.3.2 (Goudet 1995). Ne was estimated based on two different methods: the linkage disequilibrium method and the coancestry method using the software Ne estimator (Waples and Do 2008). We also cre- ated discriminant analysis of principal components (DAPC) using the R pack- age ADEGENET (Jombart 2008) and performed Bayesian clustering using STRUCTURE v.2.3.4 (Pritchard et al. 2000). To calculate population differ- entiation between the sampled locations, we calculated pair-wise FST (Weir and Cockerham 1984), Nei’s genetic distance, and IBD using SpaGeDi v1.5 (Hardy and Vekemans 2002). In the end, tests of past population bottlenecks were performed by calculating the difference between the expected heterozy- gosity and mutation-drift equilibrium heterozygosity in the software Bottle- neck (Piry et al. 1999).
2.3. Species distribution model (SDM) In order to reconstruct the ancient and present habitat range of Chinese Grouse, I used the maximum-entropy approach (MaxEnt, ver. 3.3.3k) (Phillips et al. 2006) based on climate data during the current, mid-Holocene (6000 years BP), last glacial maximum (LGM, 21,000 years BP), and last interglacial (LIG, 120,000–140000 years BP) in chapter 2. The 19 bio-climatic variables from the WorldClim 1.4 database were at a spatial resolution of 2.5 arc- minutes (Hijmans et al. 2005). I used 41 location records, which were set in a spatial distance threshold of 0.083 decimal degrees (i.e. 5ʹ, about 8 km)
17 (Dormann et al. 2007; Lu et al. 2012). Furthermore, a 10% training presence logistic threshold was used to define the minimum probability of suitable hab- itat. A GAP analysis (Scott et al. 1993) between current distribution and co- niferous forests were performed to assess conservation status. All data prepa- ration and raster calculations were performed in ArcGIS 10.5.
2.4. Population genomic analysis Demographic history inference using PSMC In chapter 2 and chapter 4, I reconstructed the demographic history of Chi- nese Grouse and Hazel Grouse using the Pairwise Sequentially Markovian Coalescent (PSMC 0.6.5) (Li and Durbin 2011) model to infer the effective population sizes (Ne) of the sibling species across genome sequences with SNP sites. To run the PSMC method, two parameters are needed: the genera- tion time (2.5 years) and the mutation rate per generation (0.3×10-8 per nucle- otide per year) of Chinese Grouse and Hazel Grouse. The mutation rate (per nucleotide per year, μ) for Chinese Grouse and Hazel Grouse were selected from studies of willow grouse (0.299×10-8) and rock ptarmigan (0.310×10-8) (Kozma 2016). PSMC was run for 100 iterations using an initial h=q ratio of 5 and the default time patterning. Bootstrapping was performed as in Li and Durbin (2011) by resampling 500000 bp segments of the genome with re- placement to perform 100 bootstrap replicates.
Genetic diversity, FST, and Linkage disequilibrium (LD) Nucleotide diversity (π) and Tajima’s D of each location were calculated on all SNPs in unrelated individuals per site using VCFtools (0.1.14) (Danecek et al. 2011). I ran EIGENSOFT (v6.0.1) (Patterson et al. 2006) to estimate FST between sampling locations in the two species. Linkage disequilibrium (LD) was calculated as pairwise R2 for each chromosome using the R package “snpstats” (Clayton 2015). R2 values were plotted against genetic distance be- tween each pair of SNPs in order to compare the decay of LD among grouse populations. In order to determine the critical value for R2, R2 was calculated for each pair of SNPs from different chromosomes. The 99th quantile of all unlinked R2 values was considered as the baseline beyond which genetic link- age is likely to cause LD (Chapter 3).
Phylogenetic relationships, PCA, population structure, and introgression To estimate phylogenetic relationships, a neighbour-joining (NJ) tree with 100 bootstrap replicates was inferred using all SNPs from all samples (Chapter 4). The split between Hazel and Chinese grouse was estimated to 1.74 Mya as- suming a molecular clock. Principal component analysis (PCA) was carried out without calling genotypes following the procedure outlined in (Fumagalli
18 et al. 2013) (Chapter 3). The population structure was inferred by admixture (v1.3) (Alexander et al. 2009) with a maximum-likelihood approach. To ex- plore genetic divergence among all individuals, the pre-defined genetic clus- ters (K) were set from 2 to 4 with 10,000 EM iterations each run. We per- formed clustering analysis via the maximum-likelihood approach, imple- mented in Admixture, assuming 2 – 4 ancestral populations (K=2 – 10). The lowest cross-validation (CV) error was calculated for each model with nine modeled clusters. The clustering results (K=2 – 4) were then visualized using R. (Chapter 3) To estimate introgression ABBA-BABA analyses were performed with the ANGSD toolbox (Korneliussen et al. 2014) (analysis of next-generation se- quencing, version 0.930) using Lagopus lagopus as an outgroup species. We ran ABBA-BABA analyses and calculated the D statistics for analysis of all combinations of Chinese Grouse and Hazel Grouse populations. To determine the significance of the D statistics we calculated Z scores for 1-Mb blocks using jackknifing (Reich et al. 2009) with an R script from the ANGSD toolbox (Korneliussen et al. 2014). (Chapter 3).
Local adaptation, inbreeding, and genetic load To identify genome-wide selective sweeps associated with high altitude adap- tation, I calculated the average pooled heterozygosity (HP) and the genome- wide distribution of the fixation index (FST) for Hazel Grouse and Chinese Grouse using a sliding-window approach, which involved 100 kb windows with 50 kb increments (Chapter 3). To explore evolution within the functional gene ontology catalogue, we used Kobas (Xie et al. 2011) to annotate the se- lected genes of the Chinese Grouse with the chicken genome (GRCg6a), which were submitted to Gene Ontology and KEGG databases for enrichment analysis. Then I used a false discovery rate (FDR) corrected binomial distri- bution probability approach to test significant enrichment gene function at a level of P < 0.05 (Benjamini and Hochberg 1995). In order to estimate inbreeding in different populations of the two sibling species, runs of homozygosity (ROH) were called using BCFtools/RoH (Narasimhan et al. 2016) from next-generation sequencing data. The sum and length of ROH in Chinese Grouse and Hazel Grouse were measured in each individual (Chapter 5). To estimate genetic load in the Chinese and Hazel grouse, I used the mappings and genome annotation of the Chicken genome. The variant effect predictor tool (McLaren et al. 2016) was used to identify loss-of-function mutations (transcript ablation, splice donor variant, splice ac- ceptor variant, stop gained, frameshift variant, in frame insertion, in frame deletion, and splice region variant), missense, and synonymous mutations on the filtered SNP calls. SIFT was then used to identify those missense muta- tions which are likely deleterious (Ng and Henikoff 2003). As an indication of mutational load, for each individual, we counted the number of genes con-
19 taining one or more loss-of-function and the total number of deleterious mis- sense mutations divided by the number of synonymous mutations (dN/dS) (Fay et al. 2001). We excluded all missense mutations within genes containing a loss-of-function mutation, as these are expected to behave neutrally. Divid- ing by the number of synonymous mutations mitigates species-specific biases, such as mapping bias due to the fact that the reference genome was derived from the chicken, coverage differences, and mutation rate.
20 Results and discussion
3.1. Genetic differentiation in Sichuan jay and Siberian Jay (Chapter 1) The purpose of this chapter was to use microsatellite markers to estimate ge- netic variation and population differentiation of different Sichuan Jay and Si- berian Jay populations across the QTP and Northern Eurasia. The results showed that the Sichuan jay, which inhabit a narrow band of fragmented high elevation coniferous forests, has similar levels of genetic variability as com- pared to the Siberian Jay since estimates of genetic diversity were similar when comparing values of AR, Ho and He. Based on the 9 common microsat- ellites markers, I found heterozygosity and allelic richness to be of the same magnitude in both species. Geographic structure and separation among the sampled Sichuan Jay was supported by the discriminant analysis of principle components (DAPC) which shows three main clusters among these samples based on the Bayesian Information Criterion (Figure 2). These results were confirmed by STRUC- TURE analyses which support three weakly differentiated clusters when sorted by assignment probability (Q). For all jay samples, the STRUCTURE analysis revealed three distinct clusters K=3 (delta-K) supported by the Evanno method (Evanno et al. 2005). A highly significant correlation was found between genetic differentiation and geographic distance of populations when all populations were included in the IBD analyses (r2 = 0.5113, P < 0.001). When IBD analysis was run only on Siberian jay populations, the correlation was also highly significant (r2 = 0.8568, P < 0.001). In addition, I found evidence of mutation-drift heterozy- gote excess (Heq) in all populations except when assuming a strict step-wise mutation model (SMM). I therefore conclude that there is a strong indication of past population bottlenecks in both species.
21
Figure 2. Discriminant analysis of principle components (DAPC) scatterplot. The three DAPC clusters are based on 11 microsatellite markers from 58 P. infaustus from Sichuan, China.
The described situation should be similar in other boreal birds in Siberia and the QTP. Some studies have suggested that many of these populations, like hazel grouse, might come from one or a few refugial lineages from central and western coniferous forest in Eurasia, which might have been almost com- pletely eradicated during the major glaciations (Baba et al. 2002; Englbrecht et al. 2000). From a conservation perspective, preserving boreal forest sur- rounding the QTP would also benefit the QTP-residing bird species of genera such as: Tetrastes, Strix, Aegolius and Picoides. The dispersal abilities of different species and their response to the ice age and subsequent warming correspond to their divergence and speciation (Barluenga and Meyer 2005; Hewitt 1999). Large-scale comparative popula- tion genetic studies of sister-species, like Sichuan jay and Siberian jay, of bo- real coniferous forest would provide a comprehensive view of the genetic con- sequences of the glaciation events on species with different life histories, and their effect on the current genetic diversity and geographical distribution pat- terns.
22 3.2. Genomic analysis of demographic history and ecological niche modeling in the endangered Chinese Grouse (Chapter 2) In chapter 2, I used whole genome sequencing data from Chinese Grouse to explore the evolutionary history and conservation genomics of this species from the QTP. From PSMC, I found fluctuations in Ne throughout the Pleis- tocene. From the beginning of the simulations to the last interglacial period 4 (LIG, 120–140 kya), Ne of Chinese Grouse was at a high number (20×10 ). Ne reached a peak before the last glaciation maximum (LGM; 40-100 kya ago). Since a significant decrease has happened during the LGM (~20 kya ago), when substantial alpine glaciations (such as Gongga glacial II) would likely have resulted in extensive loss of conifer forests. The population expansion can be explained by the warmer weather during the Greatest Lake Period (30,000 - 40,000 years ago) as the alpine conifer forests, the primary habitat for Chinese Grouse (Sun et al. 2003), reached their greatest extent at this time (Zhan et al. 2011; Zhao et al. 2013). The population decline during the LGM, when substantial alpine glaciations is likely explained by a presumed exten- sive loss of coniferous forests (Zheng et al. 2002). To assess the role of climate change, I used SDMs and palaeoclimatic data on temperature and precipitation to estimate the suitable habitat of Chinese Grouse during the time periods of 120-140 kya (LIG), 21 kya (LGM), 6 kya (Mid-Holocene), and the present. The model revealed that suitable habitat dur- ing the LGM was more widely available than both during the warmer LIG and mid Holocene. Climate events during the LGM period resulted in an expan- sion of suitable habitat to the low altitude region in eastward areas compared with LIG period and a shrink from this low altitude region to a more westward distribution during the mid-Holocene period. The result for the Chinese Grouse is congruent with the patterns of population size change estimated from the PSMC model, which suggested an expansion in population size from LIG period and then a peak and a bottleneck occurring at the LGM. Signifi- cantly, the suitable habitat of Chinese Grouse expanded toward lower altitude in the eastern areas during the LGM; in contrast, the population moved into high altitudes in the western areas during the mid-Holocene. It is interesting that the geographic distribution of Chinese Grouse populations shifted in an east-west direction from LIG to the LGM and the LGM to mid-Holocene (Fig- ure 3). The reason is probably that the shift of coniferous forests was signifi- cantly affected by seasonal extremes in temperature. In addition, the loess plateau stopped the forest from spreading to the north.
23
Figure 3. The change in predicted distributions of the Chinese Grouse: a) from LIG to LGM; b) from LGM to Mid-Holocene; c) from Mid-Holocene to Present-day. Map data was from WorldClim-Global Climate Data, which is free data for ecological modeling and GIS.
Like previous work, my results suggest that the QTP had several Pleistocene refugia which have affected Ne in this grouse inhabiting coniferous forests (Zhan et al. 2011). Narrow high altitude boreal forests were preserved in the southeast edge of the QTP by the uplift which comprise one of the key high- altitude biodiversity hotspots in the world (Myers et al. 2000). This region harbors one of the world’s richest fauna and floras, because it harbor large and complicated refugial areas in different mountain regions, lowland, rivers and basins (Ursenbacher et al. 2006; Xing and Ree 2017). The Chinese Grouse distribution area under a climate change scenarios is similar to other montane species (Pounds et al. 1999), which would shift north- ward and to a higher altitude (Lyu and Sun 2014). As suggested by my anal- yses, the long-term fragmentation of coniferous forests should have had an effect on gene flow, speciation and divergence of these endemic isolated pop- ulations.
24 3.3. Conservation genomics in the boreal forest: Population structure and local adaptation in Chinese Grouse and Hazel Grouse (Chapter 3) Chapter 2 only focused on the demographic history of the Chinese Grouse using de novo whole genome sequencing and SDM modelling. From Chapter 3 and onwards, I used 29 resequenced individuals to explore conservation genomics in the boreal forest sibling species Chinese Grouse and Hazel Grouse found across Eurasian boreal forests and the QTP, with 3 ptarmigan samples as outgroup. In chapter 3, I first determined population structure and looked for signs of local adaptation using the sequenced genomes from the grouse species. The analyses here provide insights into the population struc- ture and levels of diversity in these two species. Furthermore, ABBA-BABA tests within the two species provides evidence of introgression by gene flow occurring between sub-populations of each species. Through selection anal- yses using the sibling grouse species, I detected that the Chinese Grouse in- habiting the QTP high altitude environment show evidence of having evolved adaptations to hypobaric hypoxia and high ultraviolet radiation. Nucleotide diversity, π (Nei and Li 1979), showed higher diversity in Hazel grouse compared to Chinese grouse, not surprisingly since species range and effective population size are often correlated (Frankham 1995a; Groombridge et al. 2000). Within species, the isolated Chinese Grouse population in Qilian Mountains and in Scandinavia (SWE) of Hazel grouse had the lowest diver- sity. PCA and admixture analyses showed strong population genetic structure within both species, which is as expected. In Chinese Grouse, the QLS popu- lation is special since it was clearly isolated from the other two populations. My analysis confirmed strong separation of the QLS population previously suggested from ecological niche modelling (Lu et al. 2012; Lyu and Sun 2014). There was also evidence for structure within Hazel Grouse populations as suggested by PCA and Admixture analyses. FST –analyses also showed strong population difference among the three populations, especially between the Swedish and German populations. This suggest Swedish Hazel Grouse have become isolated and lost genetic diversity during the recolonization of the bo- real forests in Scandinavia since the last glaciation: more than populations in central Europe. The Chinese population of hazel grouse displayed the highest genetic diversity, not surprisingly since the Far East was a refugium for hazel grouse during the last glacial maximum. Chinese Grouse and Hazel Grouse both reside in extremely cold forests during winter, either in high-altitude coniferous forest in the QTP or boreal forest in the Eurasian sub-arctic. The environmental conditions in the Tibetan plateau area are in some ways more extreme than in the Eurasian boreal forests
25 when considering hypobaric hypoxia and high solar radiation. In this chapter, using FST and θπ methods, I identified 263 genes with top 5% maximum FST values and top 5% maximum θπ1/θπ2 values suggesting divergent selection in Chinese Grouse (Figure 4). The majority of previous studies on high-alti- tude adaptation have focused on species that dwell above the tree line on rocky steppes and grassland of the QTP, such as ground tits (Qu et al. 2013), yak (Qiu et al. 2012), grey wolf (Zhang et al. 2014), humans (Lorenzo et al. 2014) and Tibetan migratory locusts (Ding et al. 2018). Here we show high altitude molecular adaptations in a species which reside below the tree line. Thus, the Chinese Grouse genome has been shaped by adaptation to the high-altitude habitat compared to the old-adapted Hazel Grouse.
Figure 4. Analysis of genes under local adaptation. Genomic region (red points) lo- cated at top right of the dashed lines are candidates of high-altitude adaptations.
In conservation and management of species, molecular data provide an excel- lent tool to study populations and enable us to identify the distribution of gen- otypes as well as pattern of isolation. Such knowledge is especially important
26 for high latitude and high altitude species, since these species are very sensi- tive to demographic events and ongoing climate warming (Fedorov et al. 2020; Höglund 2009). My results provide insights into the strong population structure in each of the sibling species as well as provide evidence of local adaptation of Chinese Grouse. My analysis also identified certain regions (SWE and QLS) with particularly low diversity, which could be used to set priorities for conservation efforts.
27 3.4. Demographics and divergence of Chinese Grouse and Hazel Grouse (Chapter 4) To provide new insights into the role of recent ice ages in species demographic divergence and population history in southern Eurasian boreal forests, I used 29 re-sequenced genomes of the sibling species Chinese Grouse and Hazel Grouse to again run PSMC. I constructed a Neighbor-Joining (NJ) tree based on genome-wide data, which showed clear species and population structure. An analysis of full diploid genomes from the sibling species using PSMC modelling showed very different population demographic histories between the sibling species, such that peaks and bottlenecks of effective population size occurred at different times in the two species. For the Chinese Grouse, the northern population QLS experienced persistently low effective popula- tion sizes and have declined since 0.5 Mya and the southern population, LHS and ZN, showed bottlenecks 2 million years ago and 17,000 years ago and population peak at 30,000 – 17,000 years ago (Figure. 5). For the Hazel Grouse, the population peaked at 400 kya, after which the population trends differed between the Chinese and European Hazel Grouse populations. All the European populations appeared to have undergone persistent declines in pop- ulation size after the peak in the last glacial period (LGP). In contrast, effective population size of the Chinese Hazel Grouse showed an increase to a large population size during the LGM, and thus consistently higher population sizes than European Hazel Grouse. In the Pliocene period (5.333-2.58 Mya), the climate was similar to the pre- sent (Robinson et al. 2008); a cool, dry, seasonal climate, which had consid- erable impacts on Pliocene vegetation. All grouse species occurred within the temperate, boreal, and Arctic biogeographical zones of the Northern Hemi- sphere and are adapted to cold climates (Johnsgard 1983). In the beginning of the Quaternary, with cyclic growth and decay of continental ice sheets (Denton et al. 2010; Schlee 1985), the distribution patterns of boreal forests and steppes changed worldwide. These boreal forests eventually supported forest birds such as Hazel Grouse and Black Grouse (Tetrao tetrix), which are cold-adapted species. The uplifting of the QTP and the continuous accumula- tion of loess in the Loess Plateau was the main driving factor to limit the dis- tribution of boreal forest in China and the split of the sibling species. It is hypothesized that the extinct Bonasa praebonasia population, the common ancestor of the sibling species Chinese Grouse and Hazel Grouse, spread throughout the existing boreal forests from Northern Eurasia and across the Loess Plateau into the QTP (Liu 1994; Storch 2000; Sun et al. 2003). My re- sults support that the divergence of the ancestor of Chinese Grouse and Hazel Grouse was synchronized with the formation of the Loess Plateau, when the boreal forest in Northern Eurasia and the QTP became disconnected.
28
Figure 5. Historical changes of Chinese Grouse and Hazel Grouse in effective popu- lation size from PSMC analysis of whole-genome sequences. Profiles labeled JHGO (green) are from Hazel grouse, ZN (olive), LHS (blue) and QLS (red) are from Chi- nese grouse.
Additionally, the Multiple Sequentially Markovian (MSMC) analysis of the sibling grouse species from ~ 20,000 ybp to the present suggested that popu- lation sizes have decreased constantly from 20,000 years ago to 200 years ago. The dramatic decrease in Ne from 10,000 years ago to the present, coincide with the climate change in the Holocene (10,000 years ago to 5000 years ago) and increasing human forest exploitation from 8,500 years ago (Mygatt 2006). In the early-Holocene around 10,000years ago, the interglacial period was firmly established, exhibiting a warmer and moister climate than today. The ice cover regressed and trees were able to re-colonize the northern taiga. How- ever, at the beginning of the Neolithic Revolution (8,500 years ago), humans started to exploit forests not only for wood and food, but also for space (Perlin 2005; Um et al. 2009). Nowadays, Boreal ecosystems world-wide are threat- ened by direct human activity and climate change (Gauthier et al. 2015).
29 3.5. Purifying selection in Grouse is more efficient in large populations (Chapter 5)
In chapter 4, the MSMC analysis of the two grouse species showed that Ne have decreased constantly from 20,000 years ago to 200 years ago. To infer the effects of the population size on inbreeding levels, I took a comparative genomic approach to compare inbreeding depression, purifying selection and genetic load between the Hazel Grouse from a large distribution and the Chi- nese Grouse with a narrow distribution, as well as Willow Ptarmigan and Rock Ptarmigan, which both display wide ranges and large population sizes. I measured individual inbreeding as the proportion of the genomes that were in ROH (FROH) identified from the whole-genome resequencing data. The ROH of populations is related to their effective population size, with smaller populations tending to have more ROHs and large populations fewer ROHs (Ceballos et al. 2018). In my study, the sum of ROH in Mbs showed that the Chinese Grouse had lower inbreeding levels than Hazel Grouse. How- ever, the European populations of Hazel Grouse and QLS population of Chi- nese Grouse were significantly more inbred within species. There was a large range of FROH in the populations of Chinese Grouse and Hazel Grouse. FROH measured using ROH ranged from 0.02 to 0.24 among Chinese Grouse populations and from 0.01 to 0.43 among Hazel Grouse pop- ulations. The QLS population of Chinese Grouse and European population of Hazel Grouse (including SWE, GER and NEP) were both inbred. The QLS population has experienced persistent low effective population size in a small refugium in northeast edge of the QTP, the Qilian Mountains (Figure 5). More ROHs and larger FROH provide evidence that the QLS population have expe- rienced long time of inbreeding but survived for thousands of generations at small effective population size (Charlesworth and Willis 2009; Keller and Waller 2002). Ne in the European populations of Hazel Grouse was affected many times by the ice ages before 20 Kya which led to population fragmenta- tion and isolation, especially in Sweden and Germany (Ingolfsson et al. 2006). In contrast, the XJL population was probably outbred with a large population size, and the area was a refugium during the last ice age (Abbott et al. 2000; Saarnisto and Lunkka 2004) To test whether inbreeding and low effective population size led to an ac- cumulation of deleterious mutations in Chinese Grouse and Hazel Grouse, I characterized genetic load. By comparing the ratio between deleterious mis- sense mutations and synonymous mutations. Purging can reduce the inbreed- ing load and the depression of fitness through increased purifying selection facilitated by inbreeding (García-Dorado 2012; Hedrick 1994; Tallmon et al. 2004). Identifying the efficiency of purifying selection in large or small pop- ulations of endangered species could aid in animal conservation (van der Valk et al. 2019). In my study, genome-wide measures of genetic load, measured as the ratio between mildly deleterious missense mutations and synonymous
30 mutations, showed that purifying selections is more efficient in Hazel Grouse, which has a larger population compared to Chinese Grouse. However, when load was measured as the ratio between highly deleterious loss-of-function mutations and synonymous mutations, purifying selection showed no large difference between Chinese Grouse and Hazel Grouse. The finding supports that purifying selection on mildly deleterious missense mutations is more ef- ficient in large populations (van der Valk 2019). Thus, in isolated populations with long-term small population size, purifying selection is less efficient (Kutschera et al. 2020). However, there was no difference when comparing highly deleterious loss-of-function alleles among the species, which means that highly deleterious alleles are efficiently purged from also the smaller pop- ulations (van der Valk et. al 2019). However, it appears as such purging has been efficient in the ptarmigan samples. Ptarmigan species are signified by huge ranges and large effective sizes (Berlin et al. 2008).
Figure 6. (A) The ratio between deleterious missense mutations and synonymous mutations. (B) The ratio between loss-of-function and synonymous mutations.
31 Conclusions and future perspective
In my thesis, I have focused on selection, demographic divergence history and conservation genomics of sibling boreal forest birds from the QTP and across Northern Eurasia. By studying these two pairs of sibling boreal forest species at the microsatellite level and whole genome level, I could show genetic dif- ferentiation, genomic diversity, demographic history, inbreeding depression and selection acting on the species and populations. This provide new per- spectives for the conservation of boreal forest species. Genetic differentiation at the microsatellite levels studied in chapter 1 showed similar levels of genetic variability between Sichuan Jay and Siberian Jay. As the first study describing the genetic diversity of Sichuan Jay, which inhabit a narrow band of fragmented high-elevation coniferous forests, the ob- served genetic differentiation confirm previous results showing that remaining Sichuan jay habitats are at present isolated and fragmented because of inten- sive forest cutting of the virgin forest during the past decades. Additionally, in all these six populations, my estimates of Ne from two methods provided reliable values, which can be verified by census data from 2000 to 2004 in Sichuan Jay. Nei’s genetic distance analysis and IBD pattern found strong di- vergence not only in the total locations of the two jays, but also among Swe- dish and Russian locations of the Siberian jay. Unfortunately, the x-y coordi- nates of the Sichuan jay samples had been lost since the date of capture and I can thus not test if the detected substructure is due to geographically separated populations. For a conservation perspective, there is an urgent need to study other populations of Sichuan jay in China and central Russian populations of Siberian jay not included in this study, to better understand the conservation value, needs and importance of each population. From chapter 2 to chapter 5, I explored genomic variation, demographic divergence and selection at the whole genome level. From demographic anal- yses of Chinese Grouse, I found that the Chinese Grouse have experienced substantial population size changes from the beginning to the LIG and reached a peak before the LGM. A high Ne persisted from Early Pleistocene to Mid Pleistocene and a population expansion occurred after the retreat of the Penul- timate Glaciation. A sharp decrease and bottleneck happened during the LGM, when the conifer forests were subjected to extensive loss. The results inferred from the whole genome sequencing and species distribution models both sup- port a history of population fluctuations. The distribution of the Chinese
32 Grouse was strongly dependent on the boreal forest cover. To protect the frag- mented boreal forest is an essential action to protect the Chinese Grouse and other species. I also used population genomics to study conservation genomics, demo- graphic divergence, local adaptation, and inbreeding of the sibling species. Chapter 3 provides the first description of population structure and genetic diversity of Chinese Grouse and Hazel Grouse from whole-genome data in Eurasia. I observed strong evidence for population structure within both spe- cies using PCA and Admixture. In Chinese Grouse, the QLS population is special since it was clearly isolated with the lowest genetic diversity, higher pairwise FST and higher LD decay compared the other two groups. In Hazel Grouse, there were strong population differences among the three populations, especially among the Swedish and German populations. The Swedish popula- tion likely became isolated and lost genetic diversity during the recolonization of the boreal forests in Scandinavia since the last glaciation. From local adap- tation analyses, I investigated the genetic mechanisms for adaptation to high- altitude adaptation using closely related sibling species. In chapter 4, I explored demographic divergence of Chinese Grouse and Hazel Grouse by running PSMC with whole genome data from 29 rese- quenced individuals. The estimated divergence time of the two Tetrastes spe- cies was found to be around 1.74 Mya. The uplift of the QTP and the contin- uous accumulation of loess on the Loess Plateau was the main driving factor to limit the distribution of boreal forest in China and the split of the sibling species before that period. The PSMC results showed very different popula- tion demographic histories between the sibling species. The QLS population of Chinese Grouse has a demographic history different from those in the south, probably as a consequence of isolation and low effective population sizes throughout their history, especially in the LGP, which contrasts to the popu- lations residing further south. The Ne of Hazel Grouse peaked at 400 kya, after which the population trends differed between the Chinese and European Hazel Grouse populations. All the European populations appeared to have under- gone persistent declines in population size after the peak in the last glacial period (LGP). Unlike species that have wide distributions or greater dispersal capabilities, the Chinese Grouse had limited contact with Hazel Grouse and the treeless Loess Plateau acted as a barrier to gene flow to the northeast of the QTP. In chapter 5, I took a comparative genomic approach to compare inbreed- ing, purifying selection and genetic load between the Hazel Grouse with a large distribution and the Chinese Grouse with a narrow distribution. The ROH results showed that both the European populations of Hazel Grouse and QLS population of Chinese Grouse were significantly inbred when I compared the ROH in different populations of these two species. The Ne of European populations of Hazel Grouse decreased during the last glaciation before 20
33 Kya, which led to fragmented and isolated populations, which had to recolo- nize North East Europe. The QLS population has experienced persistent low Ne in a small refugium at the northeast edge of the QTP, the Qilian Mountain. In contrast, the XJL population reside in an area, which was a refugium during the last ice age, resulting in a present relatively large outbred population. The genetic load analyses showed that purifying selections has been more efficient in Hazel Grouse, a species with a larger range and population size compared to Chinese Grouse. However, when measured as the ratio between LoF and synonymous mutations, the purifying selection seemed to have no large effect when comparing Chinese Grouse and Hazel Grouse. In future studies, samples from more locations are needed to increase the resolution of the results, which will make them more reliable for the whole distribution area. More Hazel Grouse samples from central Russia and the Southern Alps can help us to further understanding the demographic and evo- lutionary history of this species. More Chinese Grouse samples from southern ZN and Tibet are important for understanding subpopulation structure and aid conservation. It is also necessary to explore the fragmentation and quality of boreal forests, especially in Europe and the QTP. Many species with low dis- persal capabilities are sensitive to forest quality and connectivity, and may become isolated by unsuitable habitat without attention. Furthermore, we can use comparative genomic methods in many other genera harboring sibling species such as Strix, Aegolius and Picoides to explore how divergence and speciation history is influenced by the ice ages and the separation of the boreal taiga from the QTP. These sibling species, like the study species of this thesis, show similar distribution patterns but different dispersal capabilities. Further exploration of local adaptation to the high altitudes in the QTP and comparing with the cold and long night environment in Northern Eurasia will help under- standing genomic and life history evolution induced by the different environ- ments. Such knowledge is important for designing sustainable strategies for species conservation.
34 Svensk sammanfattning
I den här avhandlingen har jag använt mig av två par systerarter av boreala skogsfåglar för att studera divergens, selektion, demografi och bevarandebio- logi i norra Eurasien och Tibetanska högplatån på mikrosatellitnivå i kapitel 1 och på genomnivå i kapitlen 2, 3, 4 och 5. I kapitel 1, som är den första studien som beskriver den genetiska mångfalden hos sotlavskrika (Perisoreus inter- nigrans) och lavskrika (P. infaustus), använde jag mikrosatellit-markörer för att beräkna den genetiska differentieringen mellan olika populationer av sot- lavskrikor och lavskrikor. Resultaten visar på liknande nivåer av genetisk va- riation, populationsstruktur och hög genetisk differentiering både mellan ar- terna och mellan olika populationer. I kapitel 2 använde jag demografiska analyser för att undersöka föränd- ringar i populationsstorlek hos kinesisk järpe (Tetrastes sewerzowi) och fann att den har haft en stor ökning i populationsstorlek från början av den senaste istiden fram till strax innan inlandsisens maximala utbredning. Resultat från helgenomssekvensring och artdistributionsmodeller stöder att järpen gått ige- nom stora variationer i populationsstorlek. I kapitel 3 till 5, har jag använt resekvensering av genomet hos 29 individer av kinesisk järpe och järpe (T. bonasia). Med hjälp av populationsgenomiska metoder har jag utforskat genomisk variation, demografisk divergens, lokal anpassning och inavel hos de båda arterna. Jag har hittat starka bevis för po- pulationsstruktur, historiska förändringar i demografi, olika nivåer av inavel och genetisk belastning hos båda arterna. Det var stora populationsskillnader och inavelsnivåer bland de tre stude- rade populationerna av järpe, speciellt mellan de svenska och tyska populat- ionerna. Under återkoloniseringen av de boreala skogarna efter den senaste istiden har järparna förmodligen tappat genetisk variation. En population i norra delen av den kinesiska järpens utbredningsområde hade minst genetisk variation, högt parvist FST, hög koppling mellan gener, högre inavelsnivåer och högre genetisk belastning jämfört med de två andra studerade population- erna. Även järpe och kinesisk järpe skiljer sig åt när de analyserades med avse- ende på av genetisk belastning. Hos järpen har utrensning av svagt skadliga mutationer har fungerat mer effektivt jämfört med den kinesiska järpen, ef- tersom järpen har en större populationsstorlek och större utbredningsområde. Däremot, när jag jämförde genetisk belastning som ett förhållande av mycket
35 skadliga mutationer, som ger upphov till funktionsbortfall, och synonyma mu- tationer fanns det inga större skillnader mellan de båda arterna. Mina fynd visar att små, isolerade och fragmenterade populationer av skogsfåglar tappar genetisk variation och därmed kan vara sårbara för framtida utmaningar samt att olika populationer kan användas för att spåra tidigare förändringar i livs- miljöer och hur de anpassat sig till lokala förhållanden.
36 中文结结
青藏高原高山针叶林和欧亚大陆的亚寒带针叶林动物群中有很多姐们 种以及分布不连续的亚种群。我主要利用青藏高原高山针叶林的特有 鸟类斑尾榛鸡和黑头噪鸦和生活在北方针叶林的姐妹种花尾榛鸡和北 噪鸦进行这一动物群的种群分化、种群动态和保护研究。我的研究主 要从微卫星水平（第一章）和全基因组水平（第二章到第五章）两个 方面进行。 在第一章里，我用微卫星的方法对黑头噪鸦和北噪鸦种群进行遗传差 异的研究。这是黑头噪鸦首次在遗传多样性的方向的研究。结果显示 分布区广泛的北噪鸦和分布区狭窄的黑头噪鸦各种群之间的遗传多样 性相似。但是他们有很强的遗传结构和很大的遗传差异。同时，我还 发现黑头噪鸦和北噪鸦的各种群在历史上均出现过瓶颈效应。通过对 遗传距离的研究，我还发现不论是黑头噪鸦还是北噪鸦，各种群之间 都有很强的遗传分化。从保护的角度来说，对青藏高原东南部黑头噪 鸦种群和俄罗斯中部的北噪鸦种群的研究是非常有必要的。 从第二章到第五章我开始从基因组水平进行遗传差异、种群历史、局 部适应和近交及遗传负荷的研究。在第二章里，通过对斑尾榛鸡的种 群历史变化分析， 我发现斑尾榛鸡从起始时间（约 700 万年前）到末 次间冰期之间一直保持着很大的人口规模。其历史有效种群大小从末 次间冰期开始增长，到末次盛冰期之前到达顶峰，此时青藏高原经历 了大湖时期，森林植被分布达到峰值。由于针叶林在末次盛冰期时遭 受了大面积的损失，斑尾榛鸡的有效种群数量急剧减少并进入瓶颈 期。物种分布模型的结果可以从历史分布区的角度支持斑尾榛鸡的这 种有效种群的历史变动。模型结果显示，从末次间冰期到末次盛冰 期，斑尾榛鸡的历史分布区随着针叶林的变化向西部高海拔地区扩 散。由于斑尾榛鸡对针叶林的依赖程度很高，所以保护青藏高原地区 零散的高山针叶林对于保护斑尾榛鸡及其同域分布的其他物种如黑头 噪鸦、四川林鸮和血雉等具有重要意义。 从第三章开始，我开始从基因组学的角度来研究青藏高原高山针叶林 和欧亚大陆北方针叶林的姐们种之间的种群历史动态、种群分化、青 藏高原的局部适应、保护基因组学以及近亲繁殖和遗传负荷等。第三 章的研究是首次利用全基因组数据从遗传结构、遗传多样性、基因渗 透以及局部适应的角度对欧亚大陆针叶林姐妹种斑尾榛鸡和花尾榛鸡 的保护基因组学进行的研究。主成分分析和遗传结果显示两个物种内 有很强的遗传结构。斑尾榛鸡的祁连山种群与莲花山种群和卓尼种群
37 相比，显得很特殊。祁连山种群遗传多样性最低，并且和其他两个种 群明显的分离。在花尾榛鸡种群中，三个种群之间有很大的差异，尤 其是德国和瑞典之间存在很大差异。另外通过第斑尾榛鸡的局部适应 分析，我还发现相关的高海拔适应的遗传机制。 在第四章里，通过对 29 个斑尾榛鸡和花尾榛鸡个体的重测序数据进行 PSMC 分析，我发现了他们之间有大的历史种群差异。一方面，斑尾榛 鸡和花尾榛鸡的分化时间约为 174 万年前，这与青藏高原的隆起和黄 土高原的黄土堆积有关。在两个地质事件与北半球冰期的共同作用 下，高山针叶林从青藏高原到欧亚大陆亚寒带针叶林的连续分布从黄 土高原地区断开，这也是导致众多森林姐妹种或者物种分布区断开的 主要原因。另一方面，姐们种之间的历史种群变化差异很大。斑尾榛 鸡的祁连山种群与莲花山和卓尼种群不同，在整个种群历史中一直有 效种群一直较小。在末次冰期期间，祁连山种群开始孤立，其他两个 地区有效种群数量开始增加，并在每次盛冰期之前到达峰值。而花尾 榛鸡的有效种群数量在 40 万年前就达到了峰值，之后开始下降，并在 末次冰期时进入瓶颈期。末次盛冰期之后，两个姐妹种的有效种群军 开始下降，并一直保持较低的种群大小。 由于两个物种均在历史上有过瓶颈期，我在第五章里采用比较基因组 学的方法对分布区较大的花尾榛鸡和分布区狭窄的斑尾榛鸡进行了近 亲繁殖，纯化选择和遗传负荷的研究。结果显示欧洲的花尾榛鸡种群 和祁连山的斑尾榛鸡种群均存在较高的近交程度。末次冰期时，斑尾 榛鸡祁连山种群和花尾榛鸡的欧洲种群均出现了较低的有效种群数 量，这与他们有较高的近交程度相关。遗传负荷的研究表明，纯化选 择在分布区更广泛的花尾榛鸡中更加有效。但是，当我用功能性基因 的丧失与同义突变间的比率参考时，斑尾榛鸡和花尾榛鸡的纯化选择 似乎没有太大影响。 从长远来看，在斑尾榛鸡南部种群和花尾榛鸡的中西伯利亚种群及阿 尔卑斯山种群的研究对于两个姐们种的保护遗传学和进化历史的很有 必要。同时，对更多的相似姐们种的全基因组测序及种群基因组分 析，对于我们了解青藏高原高山针叶林和欧亚大陆北方针叶林的动物 群的分化历史非常有意义。
38 Acknowledgments
Finally, I start to write my acknowledgments. In these four years, I have achieved a lot of great things: I got married, had a cute daughter and now I almost finished my PhD. I would like to thank many people.
First, I would like to express my sincere thanks to my dear supervisor Jacob Höglund who accepted me as a doctoral student at the beginning and encour- aged me all the time. Words cannot describe how thankful I am. I feel very lucky to be your student. In these four years, you let me do what I wanted to do not only in my PhD work but also in my life. Please accept my best thanks. Thanks for everything you did for me.
Then I would like to thank my second supervisor Yue Hua Sun, who is not only an excellent scientist, but also has professional skills in football. It is hard to find words to express my gratitude. You started guiding me in scientific work and football in 2009 and has until now. You are both my best friend and master.
I also want to thank Peter Halvarsson, another second supervisor. It is great to be your first student. I think I am the first student that I has participated at my supervisor’s defense party. You instructed me in the experiments when I first started my PhD work. I am happy and lucky to be your first student. Thanks for everything.
Bing Gao, my friend in China, I appreciate your help. You are not my super- visor, but you still gave me lots of help with my genome analysis. Gunilla and Maria, thanks for your help in my lab work. You two always had time and patience for me when I asked you many questions during my lab work (I think so). I also need to thank my friend Chun Rong Mi, a silent man, who solved my questions when I need help with statistic. Many thanks to Yun Fang and Yingxin Jiang, who helped with sample collection when I needed it. You are very professional. Yun Fang, thanks for your facial mask contribution when the coronavirus epidemic raged in Sweden. Tom, thanks for your contribution in my last chapter. You are a good Ping-pong player, we need to play more in the future.
39 Per, thanks for inviting me to the many dinners and for many helping in my life in Sweden. Frank, thanks for reading and give comments on my thesis. We can discuss more about Chinese food and when you come to China we will have a feast. Also, thanks for all the fish. Many thanks to Patrik for read- ing and giving comments on my manuscripts. And to Paula, it is always lots of fun having dinner at you place. You and Patrik are my good friends.
Thanks also to Mattias, Tor’s father, who always helped me when I needed a helping hand here in Sweden. Thank you, Ivain. You are a good friend. There are lots of fun moments: movies, swimming and chili. Thanks to Johanna who helped me take pictures. Looking forward to see more pictures you take for my daughter. To Sara, Jay, Will, Josefine, Andre, Warren, for many lots of fun during our lunch conservations. Many thanks to my Ahmed and Pablo. We have many interesting conversations in our office. Thanks to Anssi, Kate- rina, Mette for good suggestions to me in our lab meeting.
I would also like to say thanks my Chinese friends, Ting Sun, Ming Zhao, Hailiang Fang, Qi, Can Wei, Lijun, who help me in my life. Many thanks for all fun outside of work with Uppsala F4: Shuangshuang Zeng, Shiyu Li, Chengyu Wen, and me. I will always remember how much fun we have to- gether: sports, drink and drunk, dancing, lunches and dinners.
To the People at Animal Ecology, thanks for everything. I am so lucky to be working in here with you. Thanks to the people in our football team of IOZ. There are many football matches when I come back to China, which have lots of fun.
I also want to thank to my parents and sisters for always supporting me in my work. Thanks to my daughter Hongdou who gives me a chance to be a good father. And in the end, I’ll forever be grateful, Tong, my love. It is a fortune to meet you! You are as warm as the sunshine in my life. I will love you forever.
宋凯 2020 年 4 月 16 日 乌普萨拉
40 Bibliography
Abbott RJ, Smith LC, Milne RI, Crawford RM, Wolff K, Balfour J (2000) Molecular analysis of plant migration and refugia in the Arctic Science 289:1343-1346 Alexander DH, Novembre J, Lange K (2009) Fast model-based estimation of ancestry in unrelated individuals Genome research 19:1655-1664 An Z (2000) The history and variability of the East Asian paleomonsoon climate Quaternary Science Reviews 19:171-187 Baba Y, Fujimaki Y, Klaus S, Butorina O, Drovetskii S, Koike H (2002) Molecular population phylogeny of the hazel grouse Bonasa bonasia in East Asia inferred from mitochondrial control-region sequences Wildlife Biology 8:251-260 Barluenga M, Meyer A (2005) Old fish in a young lake: stone loach (Pisces: Barbatula barbatula) populations in Lake Constance are genetically isolated by distance Molecular Ecology 14:1229-1239 Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing Journal of the Royal Statistical Society: series B (Methodological) 57:289-300 Berlin S, Quintela M, Höglund J (2008) A multilocus assay reveals high nucleotide diversity and limited differentiation among Scandinavian willow grouse (Lagopus lagopus) BMC Genetics 9:89 Ceballos FC, Joshi PK, Clark DW, Ramsay M, Wilson JF (2018) Runs of homozygosity: windows into population history and trait architecture Nature Reviews Genetics 19:220 Charlesworth D, Willis JH (2009) The genetics of inbreeding depression Nature Reviews Genetics 10:783-796 Clayton D (2015) snpStats: SnpMatrix and XSnpMatrix classes and methods R package version 1 Colella JP, Talbot SL, Brochmann C, Taylor EB, Hoberg EP, Cook JA (2019) Conservation Genomics in a Changing Arctic Trends in Ecology & Evolution Comins NF, Kaufmann WJ (2011) Discovering the universe. Macmillan, Danecek P et al. (2011) The variant call format and VCFtools Bioinformatics 27:2156-2158 Davis MB, Shaw RG (2001) Range shifts and adaptive responses to Quaternary Climate Change Science 292:673-679 Deng T et al. (2011) Out of Tibet: Pliocene woolly rhino suggests high-plateau origin of Ice Age megaherbivores Science 333:1285-1288 Denton GH, Anderson RF, Toggweiler J, Edwards R, Schaefer J, Putnam A (2010) The last glacial termination Science 328:1652-1656
41 Ding D, Liu G, Hou L, Gui W, Chen B, Kang L (2018) Genetic variation in PTPN1 contributes to metabolic adaptation to high-altitude hypoxia in Tibetan migratory locusts Nature Communications 9:1-12 Ding Z, Sun J, Liu D (1999) Stepwise advance of the Mu Us desert since late Pliocene: Evidence from a red clay-loess record Chinese Science Bulletin 44:1211 Dormann C et al. (2007) Methods to account for spatial autocorrelation in the analysis of species distributional data: a review Ecography 30:609-628 Englbrecht C, Freyhof J, Nolte A, Rassmann K, Schliewen U, Tautz D (2000) Phylogeography of the bullhead Cottus gobio (Pisces: Teleostei: Cottidae) suggests a pre-Pleistocene origin of the major central European populations Molecular Ecology 9:709-722 Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study Molecular ecology 14:2611-2620 Fay JC, Wyckoff GJ, Wu C-I (2001) Positive and negative selection on the human genome Genetics 158:1227-1234 Fedorov VB, Trucchi E, Goropashnaya AV, Waltari E, Whidden SE, Stenseth NC (2020) Impact of past climate warming on genomic diversity and demographic history of collared lemmings across the Eurasian Arctic Proceedings of the National Academy of Sciences 117:3026-3033 Frankham R (1995a) Conservation genetics Annual Review of Genetics 29:305-327 Frankham R (1995b) Inbreeding and extinction: a threshold effect Conservation Biology 9:792-799 Frankham R (1998) Inbreeding and extinction: island populations Conservation Biology 12:665-675 Fumagalli M, Vieira FG, Korneliussen TS, Linderoth T, Huerta-Sánchez E, Albrechtsen A, Nielsen R (2013) Quantifying population genetic differentiation from next-generation sequencing data Genetics 195:979-992 García-Dorado A (2012) Understanding and predicting the fitness decline of shrunk populations: inbreeding, purging, mutation, and standard selection Genetics 190:1461-1476 Gauthier S, Bernier P, Kuuluvainen T, Shvidenko A, Schepaschenko D (2015) Boreal forest health and global change Science 349:819-822 Goudet J (1995) FSTAT (version 1.2): a computer program to calculate F-statistics Journal of Heredity 86:485-486 Groombridge JJ, Jones CG, Bruford MW, Nichols RA (2000) ‘Ghost’alleles of the Mauritius kestrel Nature 403:616-616 Hardy OJ, Vekemans X (2002) SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels Molecular Ecology Resources 2:618-620 Hedrick PW (1994) Purging inbreeding depression and the probability of extinction: full-sib mating Heredity 73:363-372 Hedrick PW, Garcia-Dorado A (2016) Understanding inbreeding depression, purging, and genetic rescue Trends in Ecology & Evolution 31:940-952 Hewitt G (2000) The genetic legacy of the Quaternary ice ages Nature 405:907 Hewitt G (2004) Genetic consequences of climatic oscillations in the Quaternary Philosophical Transactions of the Royal Society B: Biological Sciences 359:183-195
42 Hewitt GM (1996) Some genetic consequences of ice ages, and their role in divergence and speciation Biological Journal of the Linnean Society 58:247-276 Hewitt GM (1999) Post-glacial re-colonization of European biota Biological Journal of the Linnean Society 68:87-112 Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution interpolated climate surfaces for global land areas International Journal of Climatology 25:1965-1978 Huang W, Ji H (1979) The first discovery of the triphalanga fauna in Tibet and its significance to the uplift of the plateau Chinese Science Bulletin 19:885- 888 Hung C-M, Shaner P-JL, Zink RM, Liu W-C, Chu T-C, Huang W-S, Li S-H (2014) Drastic population fluctuations explain the rapid extinction of the passenger pigeon Proceedings of the National Academy of Sciences 111:10636-10641 Höglund J (2009) Evolutionary conservation genetics. Oxford University Press, Ingolfsson O, Moller P, Lubinski D, Forman S Severnaya Zemlya, Arctic Russia: a Nucleation Area for Kara Sea Ice Sheets During the Middle to Late Quaternary. In: AGU Fall Meeting Abstracts, 2006. IUCN (2020) The IUCN Red List of Threatened Species.
43 Li J, Fang X (1999) Uplift of the Tibetan Plateau and environmental changes Chinese Science Bulletin 44:2117-2124 Li J, Zhou S, Zhao Z, Zhang J (2015) The Qingzang movement: the major uplift of the Qinghai-Tibetan Plateau Science China Earth Sciences 58:2113-2122 Li L et al. (2013) Pliocene intraspecific divergence and Plio-Pleistocene range expansions within Picea likiangensis (Lijiang spruce), a dominant forest tree of the Qinghai-Tibet Plateau Molecular Ecology 22:5237-5255 Liu D (1985) Loess and the Environment. China Ocean Press, Liu D (1996) Pedosedimentary events in loess of China and Quaternary climatic cycles Liu N (1994) Studies on the Southward Migration of Severtzov's Hazel Grouse and it's Incontinuous Distribution with the other Species of Hazel Grouse Journal of Lanzhou University Lorenzo FR et al. (2014) A genetic mechanism for Tibetan high-altitude adaptation Nat Genet 46(9) 951 Lu N, Jia C-X, Lloyd H, Sun Y-H (2012) Species-specific habitat fragmentation assessment, considering the ecological niche requirements and dispersal capability Biological Conservation 152:102-109 Lu N, Sun Y-H (2011) Population viability analysis and conservation of Chinese Grouse Bonasa sewerzowi in Lianhuashan Nature Reserve, north-west China Bird Conservation International 21:49-58 Lyu N, Sun Y-H (2014) Predicting threat of climate change to the Chinese grouse on the Qinghai—Tibet plateau Wildlife biology 20:73-82 Mays HL et al. (2018) Genomic analysis of demographic history and ecological niche modeling in the endangered sumatran rhinoceros Dicerorhinus sumatrensis Current Biology 28:70-76. e74 McLaren W et al. (2016) The ensembl variant effect predictor Genome Biology 17:122 Mulch A, Chamberlain CP (2006) Earth science: the rise and growth of Tibet Nature 439:670 Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GA, Kent J (2000) Biodiversity hotspots for conservation priorities Nature 403:853 Mygatt E (2006) World’s forests continue to shrink Earth Policy Institute, Washington Narasimhan V, Danecek P, Scally A, Xue Y, Tyler-Smith C, Durbin R (2016) BCFtools/RoH: a hidden Markov model approach for detecting autozygosity from next-generation sequencing data Bioinformatics 32:1749-1751 Nei M, Li W-H (1979) Mathematical model for studying genetic variation in terms of restriction endonucleases Proceedings of the National Academy of Sciences 76:5269-5273 Newman D, Pilson D (1997) Increased probability of extinction due to decreased genetic effective population size: experimental populations of Clarkia pulchella Evolution 51:354-362 Ng PC, Henikoff S (2003) SIFT: Predicting amino acid changes that affect protein function Nucleic Acids Research 31:3812-3814 Ouborg NJ, Pertoldi C, Loeschcke V, Bijlsma RK, Hedrick PW (2010) Conservation genetics in transition to conservation genomics Trends in Genetics 26:177- 187
44 Patterson N, Price AL, Reich D (2006) Population structure and eigenanalysis PLoS Genetics 2 Perlin J (2005) A forest journey: The story of wood and civilization. The Countryman Press, Persons NW, Hosner PA, Meiklejohn KA, Braun EL, Kimball RT (2016) Sorting out relationships among the grouse and ptarmigan using intron, mitochondrial, and ultra-conserved element sequences Molecular Phylogenetics and Evolution 98:123-132 Phillips SJ, Anderson RP, Schapire RE (2006) Maximum entropy modeling of species geographic distributions Ecological Modelling 190:231-259 Piry S, Luikart G, Cornuet J (1999) BOTTLENECK: a computer program for detecting recent reductions in the effective population size using allele frequency data Journal of Heredity 90:502-503 Pounds JA, Fogden MP, Campbell JH (1999) Biological response to climate change on a tropical mountain Nature 398:611 Primmer CR (2009) From conservation genetics to conservation genomics Annals of the New York Academy of Sciences 1162:357-368 Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data Genetics 155:945-959 Qiu Q et al. (2012) The yak genome and adaptation to life at high altitude Nature Genetics 44:946 Qu Y et al. (2013) Ground tit genome reveals avian adaptation to living at high altitudes in the Tibetan plateau Nat Commun 4:2071 doi:10.1038/ncomms3071 Reich D, Thangaraj K, Patterson N, Price AL, Singh L (2009) Reconstructing Indian population history Nature 461:489-494 Ren G et al. (2017) Genetic consequences of Quaternary climatic oscillations in the Himalayas: Primula tibetica as a case study based on restriction site- associated DNA sequencing New Phytologist 213:1500-1512 Robinson MM, Dowsett HJ, Chandler MA (2008) Pliocene role in assessing future climate impacts Eos, Transactions American Geophysical Union 89:501- 502 Rózsa J, Strand TM, Montadert M, Kozma R, Höglund J (2016) Effects of a range expansion on adaptive and neutral genetic diversity in dispersal limited Hazel grouse (Bonasa bonasia) in the French Alps Conservation Genetics 17:401-412 Saarnisto M, Lunkka JP (2004) Climate variability during the last interglacial-glacial cycle in NW Eurasia. In: Past Climate Variability through Europe and Africa. Springer, pp 443-464 Schlee JS (1985) Our changing continent. Washington: Government Printing Office Scott JM et al. (1993) Gap analysis: a geographic approach to protection of biological diversity Wildlife Monographs:3-41 Shields WM (1987) Dispersal and mating systems: investigating their causal connections Mammalian dispersal patterns: The effects of social structure on population genetics:3-24 Soltis DE, Morris AB, McLachlan JS, Manos PS, Soltis PS (2006) Comparative phylogeography of unglaciated eastern North America Molecular Ecology 15:4261-4293
45 Soulé ME (1980) Conservation biology an evolutionary ecological perspective. Sinauer Storch I (2000) Conservation status and threats to grouse worldwide: an overview Wildlife Biology 6:195-205 Sun Y-H (2000) Distribution and status of the Chinese grouse Bonasa sewerzowi Wildlife Biology 6:271-275 Sun Y, An Z (2005) Late Pliocene-Pleistocene changes in mass accumulation rates of eolian deposits on the central Chinese Loess Plateau Journal of Geophysical Research: Atmospheres 110 Sun YH, Jia CX, Fang Y (2001) The distribution and status of Sichuan Grey Jay (Perisoreus internigrans) Journal für Ornithologie 142:93-98 Sun YH, Swenson JE, Fang Y, Klaus S, Scherzinger W (2003) Population ecology of the Chinese grouse, Bonasa sewerzowi, in a fragmented landscape Biological Conservation 110:0-184 Supple MA, Shapiro B (2018) Conservation of biodiversity in the genomics era Genome Biology 19:1-12 Svendsen JI et al. (2004) Late Quaternary ice sheet history of northern Eurasia Quaternary Science Reviews 23:1229-1271 Swenson JE (1991) Status and conservation of hazel grouse in Europe Ornis Scand 22:297-298 Tallmon DA, Luikart G, Waples RS (2004) The alluring simplicity and complex reality of genetic rescue Trends in Ecology & Evolution 19:489-496 Um T, Kim K, Im Y, Lee H (2009) Forest and Human Health Sangji University 12:16 Ursenbacher S, ., Carlsson M, ., Helfer V, ., Tegelstr?M H, ., Fumagalli L, . (2006) Phylogeography and Pleistocene refugia of the adder (Vipera berus) as inferred from mitochondrial DNA sequence data Molecular Ecology 15:3425-3437 van der Valk T (2019) Genomics of population decline. Acta Universitatis Upsaliensis van der Valk T, de Manuel M, Marques-Bonet T, Guschanski K (2019) Estimates of genetic load in small populations suggest extensive purging of deleterious alleles BioRxiv:696831 Wang X, Li Q, Takeuchi GT (2016) Out of Tibet: an early sheep from the Pliocene of Tibet, Protovis himalayensis, genus and species nov.(Bovidae, Caprini), and origin of Ice Age mountain sheep Journal of Vertebrate Paleontology 36:e1169190 Wang X, Tseng ZJ, Li Q, Takeuchi GT, Xie G (2014) From ‘third pole’to north pole: a Himalayan origin for the arctic fox Proceedings of the Royal Society B: Biological Sciences 281:20140893 Waples RS, Do C (2008) LDNE: a program for estimating effective population size from data on linkage disequilibrium Molecular Ecology Resources 8:753- 756 Vaurie C (1965) The birds of the Palearctic fauna. Non-passeriformes. The Wilson Bulletin London: 1-763:340 Weir BS, Cockerham CC (1984) Estimating F-statistics for the analysis of population structure Evolution 38:1358-1370
46 Weir JT, Schluter D (2004) Ice sheets promote speciation in boreal birds Proceedings of the Royal Society of London B: Biological Sciences 271:1881-1887 Wright S (1977) Inbreeding in animals: differentiation and depression Evolution Xie C et al. (2011) KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases Nucleic Acids Research 39:W316-W322 Xing Y, Ree RH (2017) Uplift-driven diversification in the Hengduan Mountains, a temperate biodiversity hotspot Proc Natl Acad Sci U S A 114:E3444 Zhan X, Zheng Y, Wei F, Bruford MW, Jia C (2011) Molecular evidence for Pleistocene refugia at the eastern edge of the Tibetan Plateau Molecular Ecology 20:3014-3026 Zhang W et al. (2014) Hypoxia adaptations in the grey wolf (Canis lupus chanco) from Qinghai-Tibet Plateau PLoS Genetics 10 Zhao S et al. (2013) Whole-genome sequencing of giant pandas provides insights into demographic history and local adaptation Nat Genet 45:67-71 doi:10.1038/ng.2494 Zheng B, Xu Q, Shen Y (2002) The relationship between climate change and Quaternary glacial cycles on the Qinghai–Tibetan Plateau: review and speculation Quaternary International 97:93-101 Zheng H, Powell CM, An Z, Zhou J, Dong G (2000) Pliocene uplift of the northern Tibetan Plateau Geology 28:715-718
47 Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1935 Editor: The Dean of the Faculty of Science and Technology
A doctoral dissertation from the Faculty of Science and Technology, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology. (Prior to January, 2005, the series was published under the title “Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology”.)
ACTA UNIVERSITATIS UPSALIENSIS Distribution: publications.uu.se UPPSALA urn:nbn:se:uu:diva-408978 2020