Horizontal transfer of an adaptive chimeric photoreceptor from bryophytes to ferns Author(s): Fay-Wei Li, Juan Carlos Villarreal, Steven Kelly, Carl J. Rothfels, Michael Melkonian, Eftychios Frangedakis, Markus Ruhsam, Erin M. Sigel, Joshua P. Der, Jarmila Pittermann, Dylan O. Burge, Lisa Pokorny, Anders Larsson, Tao Chen, Stina Weststrand, Philip Thomas, Eric Carpenter, Yong Zhang, Zhijian Tian, Li Chen, Zhixiang Yan, Ying Zhu, Xiao Sun, Jun Wang, Dennis W. Stevenson, Barbara J. Crandall-Stotler, A. Jonathan Shaw ... Source: Proceedings of the National Academy of Sciences of the United States of America , May 6, 2014, Vol. 111, No. 18 (May 6, 2014), pp. 6672-6677 Published by: National Academy of Sciences Stable URL: https://www.jstor.org/stable/23772580

REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/23772580?seq=1&cid=pdf- reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

National Academy of Sciences is collaborating with JSTOR to digitize, preserve and extend access to Proceedings of the National Academy of Sciences of the United States of America

This content downloaded from 152.3.102.254 on Mon, 30 Nov 2020 15:40:33 UTC All use subject to https://about.jstor.org/terms Horizontal transfer of an adaptive chimeric photoreceptor from bryophytes to ferns Fay-Wei Li3'1, Juan Carlos Villarrealb, Steven Kellyc, Carl J. Rothfelsd, Michael Melkonian8, Eftychios Frangedakisc, Markus Ruhsam', Erin M. Sigel3, Joshua P. Der9,h, Jarmila Pittermann1, Dylan O. Bürge', Lisa Pokornyk, Anders Larsson1, Tao Chenm, Stina Weststrand1, Philip Thomasf, Eric Carpenter", Yong Zhang°, Zhijian Tian°, Li Chen°, Zhixiang Yan°, Ying Zhu°, Xiao Sun°, Jun Wang", Dennis W. Stevenson13, Barbara J. Crandall-Stotler9, A. Jonathan Shaw3, Michael K. Deyholos", Douglas E. Soltisr,5,t, Sean W. Graham", Michael D. Windham3, Jane A. Langdalec, Gane Ka-Shu Wongn o'v1, Sarah Mathews3", and Kathleen M. Pryer3 department of Biology, Duke University, Durham, NC 27708; Systematic Botany and Mycology, Department of Biology, University of Munich, 80638 Munich, Germany; department of Plant Sciences, University of Oxford, Oxford 0X1 3RB, United Kingdom; department of Zoology, University of British Columbia, Vancouver, BC, Canada V6T 1Z4; eBotany Department, Cologne Biocenter, University of Cologne, 50674 Cologne, Germany; fRoyal Botanic Garden Edinburgh, Edinburgh EH3 5LR, Scotland; department of Biology and hHuck Institutes of the Life Sciences, Pennsylvania State University, University Park, PA 16802; 'Department of Ecology and Evolutionary Biology, University of California, Santa Cruz, CA 95064; 'California Academy of Sciences, San Francisco, CA 94118; kReal Jardin Botânico, 28014 Madrid, Spain; 'Systematic Biology, Evolutionary Biology Centre, Uppsala University, SE-752 36 Uppsala, Sweden; mFairy Lake Botanical Garden, Shenzhen, Guangdong 518004, China; "Department of Biological Sciences, University of Alberta, Edmonton, AB, Canada T6G 2E9; °BGI-Shenzhen, Shenzhen 518083, China; pNew York Botanical Garden, Bronx, NY 10458; department of Plant Biology, Southern Illinois University, Carbondale, IL 62901; rFlorida Museum of Natural History, department of Biology, and denetics Institute, University of Florida, Gainesville, FL 32611; department of Botany, University of British Columbia, Vancouver, BC, Canada V6T 1Z4; department of Medicine, University of Alberta, Edmonton, AB, Canada T6G 2E1; and wHarvard University Herbaria, Cambridge, MA 02138

Edited by David M. Hillis, The University of Texas at Austin, Austin, TX, and approved March 21, 2014 (received for review October 29, 2013)

Ferns are well known for their shade-dwelling habits. Their ability low-light, angiosperm-dominated forest canopies (7, 8). As a re to thrive under low-light conditions has been linked to the evo sult, it has been suggested that the evolution of neochrome was lution of a novel chimeric photoreceptor—neochrome—that fuses a key innovation that conferred a phototropic advantage on ferns red-sensing phytochrome and blue-sensing phototropin modules growing under low-light conditions, facilitating their modern di into a single gene, thereby optimizing phototropic responses. De versification in the "shadow of angiosperms" (3, 7, 8). Although spite being implicated in facilitating the diversification of modern ferns, the origin of neochrome has remained a mystery. We pres potentially significant from an evolutionary standpoint, the ori ent evidence for neochrome in hornworts (a bryophyte lineage) gin of fern neochrome has remained a mystery. and demonstrate that ferns acquired neochrome from hornworts In this study we investigated the origin of neochrome by via horizontal gene transfer (HGT). Fern neochromes are nested searching for homologous sequences in 434 transcriptomes, and within hornwort neochromes in our large-scale phylogenetic recon structions of phototropin and phytochrome gene families. Diver Significance gence date estimates further support the HGT hypothesis, with fern and hornwort neochromes diverging 179 Mya, long after the Despite being one of the oldest groups of land plants, the split between the two plant lineages (at least 400 Mya). By analyz majority of living ferns resulted from a relatively recent di ing the draft genome of the hornwort Anthoceros punctatus. we versification following the rise of angiosperms. To exploit fully also discovered a previously unidentified phototropin gene that the new habitats created by angiosperm-dominated ecosystems, likely represents the ancestral lineage of the neochrome phototro ferns had to evolve novel adaptive strategies to cope with the pin module. Thus, a neochrome originating in hornworts was trans low-light conditions exerted by the angiosperm canopy. Neo ferred horizontally to ferns, where it may have played a significant chrome, an unconventional photoreceptor that allows ferns to role in the diversification of modern ferns. "see the light" better, was likely part of the solution. Surpris ingly, we discovered that fern neochrome was derived from phototropism | chloroplast movement a bryophyte lineage via horizontal gene transfer (HGT). This finding not only provides the first evidence that a plant-to-plant HGT can have a profound evolutionary impact but also has Plant ceptor growth systems and that development provide information are modulated about the by surround photore implications for the evolution of photosensory systems in plants. ing environment. Major peaks in the action spectra of these informational photoreceptors lie either in the UV-blue Author (e.g., contributions: F.-W.L. designed research; F.-W.L., S.M., and K.M.P coordinated study; F.-W.L. performed research; F.-W.L., J.C.V., S.K., C.J.R., M.M., E.F., M.R., E.M.S., cryptochromes and phototropins) or red/far-red (phytochromes) J.P.D., J.P., D.O.B., L.P., A.L., T.C., S.W., P.T., E.G. Y. Zhang, Z.T., L.C., Z.Y., Y. Zhu, X.S., light regions (1). The chimeric photoreceptor neochrome J.W., isD.W.S., a B.J.C-S., A.J.S., M.K.D., D.E.S., S.W.G., J.A.L., and G.K.-S.W. contributed new remarkable exception. It fuses red-sensing phytochrome reagents/analytic and tools; F.-W.L. and S.K. analyzed data; and F.-W.L., M.D.W., S.M., and K.M.P. wrote the paper. blue-sensing phototropin modules into a single molecule (Fig. 1A) The authors declare no conflict of interest. that mediates phototropic responses (1-4). Neochromes have a restricted occurrence in the plant tree of life and are hypothe This article is a PNAS Direct Submission. sized to have had two independent origins (5)—one in the Data green deposition: The DNA sequences reported here are deposited in the GenBank database (accession nos. KJ128382, KJ128383, KJ128384, and KJ194997-KJ195254). The IKPtranscrip alga Mougeotia scalaris and another in ferns. The possession tomes can of be accessed at www.bioinfodata.org/Blast40neKP. The sequencing reads for An neochrome may be evolutionarily advantageous, as evidenced thoceros by punctatus draft genome and Pteridium aquilinum transcriptome are deposited in the greatly enhanced phototropic responses in ferns with the National neo Center for Biotechnology Information (NCBI) Sequence Read Archive under SRA096687 and SRX423244, respectively. The assembled P. aquilinum transcriptome is de chrome (3, 4) and by its phylogenetic distribution within the posited fern in NCBI Transcriptome Shotgun Assembly under GASPOOOOOOOO. Alignments and tree lineage. The early-diverging fern orders Osmundales and files Schi are available from the Dryad Digital Repository: http://dx.doi.org/10.5061/dryad.fn2rg. zaeales do not possess neochrome (3). It has been reported 1To onlywhom correspondence may be addressed. E-mail: [email protected] or [email protected]. in Cyatheales (6) and Polypodiales (3, 6), lineages that Thismostly article contains supporting information online at www.pnas.org/lookup/suppl/doi: 10. diversified following the Cretaceous/Tertiary establishment 1073/pnas.1319929111/-/DCSupplemental. of

6672-6677 | PNAS 1 May 6,2014 | vol.111 | no. 18 www.pnas.org/cgi/doi/10.1073/pnas.1319929111

This content downloaded from 152.3.102.254 on Mon, 30 Nov 2020 15:40:33 UTC All use subject to https://about.jstor.org/terms AA phytochromeS™^^^^^^^^—^ phytochrome challenged» recently, challenged and the recently, monophyly and the of monophyly all bryophytes of all bryophytes was was proposed (11). (11). We We confirmed confirmed the presence the presence of neochrome of neochrome in in neochrome hornworts throughthrough PCR PCR and and cloning, cloning, and and isolated isolated neochrome neochrome phototropin i . - ■ - - - sequences from the genera sequences Nothoceros, from the genera Megaceros, Nothoceros, Phymatoceros, Megaceros, Phymatoceros, Phaeoceros, Paraphymatoceros,Paraphymatoceros, and and Anthoceros, Anthoceros, representing representing four four B |Angio. PHOT1 of the five hornwort orders (namely, Dendrocerotales, Phyma Gymno. PHOT 1 HGT HGT Fem neo of the five hornwort orders (namely, Dendrocerotales, Phyma \ r4 tocerotales, Notothyladales,Notothyladales, and and Anthocerotales). Anthocerotales). We Wewere were un un |Angio. Angio. PHOT2 PHOT2 — Nothoceros able to obtain adequate material of the monotypic hornwort =^ Gymno. Gymno. PHOT2 PHOT2 ^—MegacerosL_QNothoceros able to obtain adequate material of the monotypic hornwort Fern PHOT1 H T^hymafoceros '—Phymatoceros Hornwort Hornwort Leiosporoceros Leiosporoceros to test tofor test the forpresence the presence of neochrome of neochrome in Leio in Leio -Paraphymatoceros neo FernFem PHOT2PHOT2 —PhaePocye?oas°ceros -Phaeoceros sporocerotales. NE0 sporocerotales. To confirm To confirm that our that hornwort our hornwort neochrome neochrome se se — Anthoceros I Lycophyte PHOTPHOT Anthoceros quence quence data data were were indeed derived derived from from the thehornwort hornwort nuclear nuclear ge ge Nothoceros |||mossPHOTA |Moss PHOTA jt^MegaSros3 nome nomeand notand notfrom from contaminant contaminant algae algae or ferns,or ferns, we performed we performed P—P Phymatoceros Hornwort genome-walking in Nothoceros aenigniaticus to obtain flanking |||t IMoss Moss PHOTBPHOTB Lj-Par^—Phymatoceros Paraphymatoceros phot Hornwort genome-walking in Nothoceros aenigmaticus to obtain flanking HLpPahraeScSos°cems -Phaeoceros Ph PH0T genomic genomic sequences. sequences. Downstream Downstream of neochrome of neochrome we found we a found a I LiverwortLiverwort PHOTPHOT :| Anthoceros 0.10.1 substitutions/site substitutions/site pseudogene pseudogene for imidazoleglycerol-phosphate for imidazoleglycerol-phosphate dehydratase dehydratase (IGPD) (IGPD) DD and, because its sequence and, because itsis sequencemost isclosely most closely related related to to other other hornwort hornwort \ horizontalN horizontal gene IGPDgene genesIGPD genes(SI Appendix, (SI Appendix, Fig. S4), Fig. we S4), are weconfident are confident that neo that neo / transfer Hornwort NEONEO —•» , /,ransfer chrome is present chrome in is the present hornwort in the hornwort genome. genome. 1 Hornwort Hornwort PHOT PHOT — — — i-^gene —'gene fusionfusion ♦ ' retrotransposition retrotransposition NeochromeNeochrome HGT HGT from from Hornworts Hornworts to Ferns. toThe Ferns.phylogenetic The distriphylogenetic distri jjjgjj Algae PHOT '• .in.i ..... i. ■ bution bution of ofneochrome neochrome inin land land plants plants (present (present only in only hornworts in hornworts conventional PHOT with introns AlgaeAigae NEO neo conventional phot with andintrons ferns) an(jcould ferns) be explained could bybe (;)explained an ancient by origin (/) analong ancient the origin along the branch that that unites unites hornworts hornworts and tracheophytes,and tracheophytes, followed followed by by r£^Eʧjj|l| Algae PHOT losses from from lycophytes lycophytes and and seed seed plants, plants, (ii) independent (ii) independent origins origins in fernsferns and and hornworts, hornworts, or (Hi) or one(iii) orone more or instancesmore instances of HGT of HGT 10001000 750 750500 250 500 250 o mya between hornworts and between ferns. hornworts To distinguish and ferns. Toamong distinguish these among three these three possible scenarios, scenarios, we we compiled compiled comprehensive comprehensive sequence sequence align align Fig. 1. The origin of fern neochrome. (A) Neochrome is a chimeric photo Fig. 1. The origin of fern neochrome. (A) Neochrome ments is a chimeric of phototropin photo- andments phytochrome of phototropin from andacross phytochrome all land from across all land receptor inin which which the the N Nterminus terminus consists consists of aof phytochrome a phytochrome sensory sensory module module , , , , • , • , , , , plants and algae, which included the corresponding domains fused to to an an almost almost complete complete phototropin phototropin sequence sequence at the Cat terminus. the C terminus. Thick Thick j?lantSu a"d alSae' ^hlCh lnC'uded tlle corresponding domains and thinthin lines lines represent represent exons exons and and introns, introns, respectively fromrespectively hornwort (length and(length notfern toneochromes, not scale), to scale). and evaluated from thehornwort resultant and fern neochromes, and evaluated the resultant (ß)(B) DatedDated phylogeny phylogeny of of phototropin phototropin and and neochrome, neochrome, gene phylogenies. showing showing Maximum neochrome neochrome likelihood and gene Bayesian phytogenies. estimates Maximum likelihood and Bayesian estimates HGT fromfrom hornworts hornworts to toferns ferns (details (details are givenare given of in phototropin SI inAppendix, SI Appendix, and phytochrome Fig. S5). Fig. The phylogeniesS5). The ofrevealed phototropin that fern and phytochrome phytogenies revealed that fern blue, brown,brown, and and yellow yellow branches branches represent represent hornwortneochromes hornwort are phototropin, embedded phototropin, within hornworthorn horn- neochromes neochromes with are embedded within hornwort neochromes with wort neochrome,neochrome, andand fernfern neochrome,neochrome, respectively. respectively. very strong branch (C) (C) Portion support Portion (Fig. of 1of theB andthe C andvery SI Appendix, strong Figs. branch support (Fig. 1 B and C and SI Appendix, Figs, phototropin phylogeny phylogeny showing showing relationships relationships S1-S3). of fern Thisof nestedfern neochrome relationshipneochrome (Fern indicates (Fern that S1-S3). neochrome This was nested relationship indicates that neochrome was NEO), hornwort phototropin (Hornwort PHOT), and hornwort neochrome NEO), hornwort phototropin (Hornwort PHOT), transferred and horizontally hornwort from neochrome hornworts to ferns, transferred along the stem horizontally from hornworts to ferns, along the stem (Hornwort NEO), with highly supported branches thickened (details are (Hornwort NEO), with highly supported lineage branches leading to Phymatocerosthickened + Nothoceros(details 4-are Megaceros leadj (Fig. tQ phymatocems + Nothoceros + Megaceros (Fig. shown in in SI SI Appendix, Appendix, Figs. Figs. S1 andS1 andS2). (D)S2). A (D) schematic A schematic depicting depicting the origin the . ^origin rr / i t— ^ i ^ Ix ° v ° 1C, arrow, and SI Appendix, Figs. S1-S3). The alternative pos of fernfern neochrome neochrome involving involving retrotransposition retrotransposition of a phototropinof a phototropin gene (and gene (and It f °I~ hencehence the theloss of loss introns), of itsintrons), fusion with itsa phytochrome, fusion withand HGT a from phytochrome, sibilities, suggestingsuggesting and eithereither HGT anan fromancientancient verticalvertical transfertransfer of of neo neo hornwortshornworts to ferns. to ferns. ' chrome (i.e., fern and chrome hornwort (i.e., fern neochromes and hornwort were neochromes reciprocally were reciprocally monophyletic) or or an an independent independent origin origin of neochrome of neochrome (i.e., fern(i.e., fern neochromes were were monophyletic monophyletic with with either either fern fern phototropins phototropins or or 40 wholewhole or or draft draft genomes genomes of plantsof plants and algaeand algae(SI Appendix, (SI Appendix, phytochromes) phytochromes) were both rejectedwere both (P < 10~30)rejected and (P were < 10-"30) never and were never Table SI) SI) and and surprisingly surprisingly discovered discovered neochrome neochrome homologs homologs from observed from in observedthe Bayesian in theposterior Bayesian tree posteriorsamples. tree samples, hornworts (Fig. (Fig. IB and1B SIand Appendix, SI Appendix, Figs. S1-S3). Figs. Analyses S1-S3). ofAnalyses We used estimatesof We used of divergence estimates time of divergenceto assess our timeHGT to assess our HGT the hornworthornwort draft draft genome genome (Anthoceros (Anthoceros punctatus) punctatus) suggest that suggest hypothesis that further, hypothesis reasoning further, that reasoningin a case of that HGT inthe a splitcase of HGT the split neochrome originated originated in hornworts, in hornworts, independent independent from the from green between the green hornwort between and hornwortfern neochrome and fernshould neochrome be younger shouldthan be younger than algae. Large-scaleLarge-scale phylogenetic phylogenetic analyses analyses and anddivergence divergence time the time split the between split thebetween hornwort the and hornwort fern lineages and themselves.fern lineages By themselves. By estimates further further demonstrate demonstrate that fernsthat fernsacquired acquired neochrome neochromeintegrating integratingfossil calibrations fossil (SI calibrations Appendix, Table(SI Appendix, S2) with Table S2) with fromfrom hornworts hornworts via horizontal via horizontal gene transfer gene (HGT). transfer (HGT). a Bayesian a Bayesian relaxed relaxedmolecular molecular clock analysis, clock we analysis,estimated thewe estimated the divergence date date between between hornwort hornwort and and fern fern neochrome neochrome to be to be Results and Discussion ~179■T79 Mya Mya with with a 95%a 95% highest highest posterior posterior density density interval interval of of133 133 Algal Neochrome. TheThe onlyonly publishedpublished algal algal neochrome neochrome is is andfrom from 229 a aMya and (Fig. 229 IBMya and (Fig. SI Appendix, IB and SI Fig. Appendix, S5). This Fig.date S5).is far This date is far single species, Mougeotia scalatisscalaris (5).(5). WeWe identifiedidentified homologshomologs more ofrecentof more than recent published than divergence published estimatesdivergence between estimates ferns between ferns neochrome in the transcriptomes of all 10 sampled members and hornworts and hornworts (at least 400(at leastMya) 400(12) butMya) is (12)congruent but is congruentwith with of the Zygnemataceae (9), including Mougeotia, Mesotaenium, the date the estimates date estimates for the stem for branchthe stem leading branch to Phymatoceros leading to Phymatoceros Cylindrocystis, and and Zygnemopsis, Zygnemopsis, but butin no in other no other algal transcriptomesalgal transcriptomes + Nothoceros ++ NothocerosMegaceros (85-244+ Megaceros Mya) (85-244(13). The Mya) disparity (13). The in disparity in surveyed (SI(SI Appendix,Appendix, Table Table SI SIand and Fig. Fig. SI). SI). divergence divergence times rejects times therejects hypothesis the hypothesis invoking invoking multiple multiple neochrome originsorigins oror losseslosses andand reinforcesreinforces thethe HGT HGT scenario. scenario. Neochrome inin Hornworts.Hornworts. AmongAmong land land plants, plants, we we documented documented Thethe the originThe origin of land of landplant plant neochrome neochrome within within the hornwortthe hornwort lin lin occurrence of of neochrome neochrome in 25in additional25 additional fern fern species species (SI Ap- eage (SI eage Apis supportedis supported by byits itsrelationship relationship to tohornwort hornwort phototropin. phototropin. pendix, Figs. Figs. S1-S3). S1-S3). Surprisingly, Surprisingly, we also we discovered also discovered neo- TheThe neo single hornwort phototropinphototropin genegene inin thethe AnthocerosAnthoceros punc punc chrome inin hornworts,hornworts, a asmall small bryophyte bryophyte lineage lineage that that diverged diverged tatus tatus draft draft genome genome completely completely lacks lacks introns introns (Fig. (Fig. ID) andID) andthus thus early inin thethe history history of of land land plants. plants. The The exact exact branching branching order closely order closely resembles resembles the theC-terminal C-terminal end endof bothof both fern fern and andhornwort hornwort among thethe three three bryophyte bryophyte lineages lineages (hornworts, (hornworts, mosses, mosses, and neochromes. neochromes.and We Wefound found this this intron-free intron-free phototropin phototropin in all in horn all horn liverworts) is is not not resolved resolved with with certainty; certainty; previous previous analyses analyses worts worts examined examined by by using using PCR PCR on ongenomic genomic DNA DNA (SI Appendix,(SIAppendix, Fig. Fig. have suggestedsuggested thatthat hornworts hornworts are are sister sister to to vascular vascular plants plants (lyco-S2 (lyco and S2 Table and TableS3). All S3). other All other phototropins phototropins characterized characterized to date, to date, phytes, ferns,ferns, and and seed seed plants) plants) (10), (10), but but this this relationship relationship wasincluding was including those those of ferns,of ferns, contain contain more more than than 20 20introns. introns. We We

Li Liet al. et al. PNAS | May 6, 2014 | vol.111 | May no. 6, 2014 |18 vol.111 || no.6673 18

This content downloaded from 152.3.102.254 on Mon, 30 Nov 2020 15:40:33 UTC All use subject to https://about.jstor.org/terms explored whether this gene might be a partial neochrome mas HGT (21, 22), or (iii) a combination of both factors. We have querading as a phototropin by using inverse PCR to obtain the 5' some evidence suggesting recurring fern-to-fern HGT might upstream genomic region in N. aenigmaticus. Multiple stop codons have been involved. For example, we discovered neochrome were encountered upstream of the Nothoceros phototropin gene, genes from two early-diverging fern orders [Gleicheniales (Dip and there was no indication of nearby phytochrome domains. tens conjugata) and Cyatheales (Alsophila podophylla and Pla These data suggest that hornworts might not have a canonical giogyria spp.)] that likely were derived from secondary HGT phototropin gene. Instead, hornwort phototropins are most events (SI Appendix, Fig. S6, arrows). These neochromes are not closely related to fern and hornwort neochromes (Fig. 1 B and phylogenetically resolved, as would be predicted based on pub C and SI Appendix, Fig. S2), implying that they likely represent lished fern species relationships (14), but instead are nested the ancestral, retrotransposed phototropin lineage that gave rise among Polypodiales (SI Appendix, Fig. S6). Furthermore, the to neochrome through fusion with the phytochrome module split between these and other fern neochromes (81 Mya, 95% (Fig. ID). On the other hand, in the phytochrome phylogeny, highest posterior density interval: 59-106 Mya; SI Appendix, Fig. hornwort phytochromes are not resolved as sister to hornwort S5) occurred long after the estimated organismal divergence and fern neochromes (SI Appendix, Fig. S3), although there is no dates for Gleicheniales (276 Mya) and Cyatheales (223 Mya) (8), branch support for this nonmonophyletic relationship. The phy a pattern that may be explained best by fern-to-fern HGT. tochrome progenitor of neochrome therefore remains unclear. Our hypothesis of potentially recurrent HGT events within ferns is not unprecedented. In angiosperms, rampant HGTs have Recurrent Fern-to-Fern HGT. We detected an extraordinary in been documented for the mitochondrial coxl homing intron. congruence between our fern neochrome gene tree and the This intron is believed to have experienced one initial "seed published phylogeny of ferns (SI Appendix, Fig. S6) (14). By transfer" from fungi that was followed by at least 80 incidents examining the entire Bayesian posterior tree sample, we found of plant-to-plant HGT among 833 diverse angiosperm species that none of the trees resolved neochromes from the same fern (23-25). Perhaps neochrome is similarly associated with mo family as being monophyletic. This conflicting pattern is bile not elements that may have facilitated its movement across observed in other fern phylogenies based on nuclear genes species (15) boundaries. and is not seen in the hornwort neochrome tree (Fig. 1C), which perfectly mirrors the published phylogeny of hornworts Evolutionary(13). and Physiological Implications of Neochrome in Hornworts. Flere we investigate and discuss the possible causes of the Our in discovery of neochrome in hornworts is an important step congruent gene tree/species tree in ferns. toward understanding the evolution of photosensory systems Incomplete sampling of extant neochrome homologs is notin plants. In the moss Physcomitrella patens, both red and blue likely to be the explanation, because neochrome has been shown light can elicit directional chloroplast movements, and these by Southern blotting to be a single-copy gene in Adiantum capillus movements are mediated by molecular interactions between veneris (2). This result was corroborated by the cloning efforts physicallythat separate phytochrome and phototropin proteins (26). produced most of our sequence data (SI Appendix, Table S3).The hornwort neochrome represents a strikingly different strat Except for Deparia spp., in which two divergent sequences egywere for integrating these two photosensory systems, combining found (SI Appendix, Fig. S6, arrowheads), we were able to isolate them into a single, chimeric gene. Light-induced directional only a single neochrome from each fern species. chloroplast movement has not yet been observed in hornworts, Next, we investigated whether an aberrant nucleotide substi probably because their epidermal cells usually contain only one tution process may have misled the phylogenetic reconstruction. chloroplast that occupies most of the cellular space. However, For example, pervasive positive selection or variation in guanine/ nearly 50 y ago, Burr (27) documented an unusual chloroplast cytosine (GC) content can obscure true phylogenetic signal photoresponse (16 in Megaceros hornworts; she discovered that the 18), thereby causing a gene tree to be incongruent with largethe chloroplasts "contract" to form compact shapes under species tree. Using codon models for tree inference potentially strong light. Although we confirmed this phenomenon in horn can accommodate complex selection profiles by allowing differ worts (SI Appendix, Fig. S8), future studies are needed to ex ent nonsynonymous/synonymous substitution rate ratios to aminefall if neochrome is responsible for the contraction response into distinct classes (19). However, we found that incorporating and to explore other possible physiological roles. codon models did not improve the incongruence between the gene tree and species tree; the resultant tree largely matches Evolutionarythat Significance of Plant-to-Plant HGT. This study pinpoints from the nucleotide substitution model, with comparable branch the origin of land plant neochrome within the hornwort lineage support values (SI Appendix, Figs. S6 and S7). Similarly, infer and demonstrates that neochrome was transferred horizontally ences based on first + second-codon positions or only on third from hornworts to ferns. The life history of ferns may help ex codon positions also yielded topologies discordant with the plain their hypothesized susceptibility to HGT. Most land plants species tree (SI Appendix, Fig. S7). share a common sexual life cycle that alternates between a dip We then used a random effects branch-site model to infer the loid sporophyte and a haploid gametophyte; only in ferns and dynamics of positive selection across the neochrome tree (20). lycophytes are the sporophytic and gametophytic phases both Only five fern branches were identified as having experienced free-living and fully independent. Seed plants insulate their significant episodic positive selection (SI Appendix, Fig. SID), gametophytes from outside interactions with relatively impervi and the proportion of positively selected codon sites along each ous cell walls in microgametophytes and by embedding mega of these five branches is very low (<3%). These results suggest gametophytes within protective sporophyte tissues. In contrast, that positive selection operated on very few codons over a lim almost all fern gametophytes are not enclosed and grow in direct, ited number of branches. Similarly, a sliding window analysis of intimate contact with other fern and bryophyte gametophytes GC content found none of the fern sequences to be deviant in (including those of hornworts). These characteristics may facili base composition (SI Appendix, Fig. SIE). Taken together, the tate the entrance of foreign genetic elements into fern germ lines nucleotide substitution processes among fern neochromes ap (i.e., via the gamete-producing structures, the antheridia and pear to be unexceptional and are not likely to explain the in archegonia, that are exposed to the environment) (28). congruence between the gene tree and species tree. To date, most documented examples of plant-to-plant HGT We therefore hypothesized that the incongruent tree could involve mitochondrial DNA and/or parasite-host transfers (29 be the result of (i) multiple fern-to-fern HGT events, (ii) an 36); only a handful of cases include functional nuclear genes elevated gene turnover rate that may have been selected for after (34, 37, 38), and even fewer have possible adaptive implications

6674 I www.pnas.org/cgi/doi/10.1073/pnas.1319929111 Li et al.

This content downloaded from 152.3.102.254 on Mon, 30 Nov 2020 15:40:33 UTC All use subject to https://about.jstor.org/terms (39).(39). Consequently, Consequently, plant-to-plant HGT generally has been plant-to-plant over- arid contigs and contigs could couldno longer no longerbe extended.HGT be extended. The finalgenerally The assembly final assembly contained contained has been over lookedlooked as a potentially assignificant a factorpotentially in plant evolution. a sin9le conti9significant a single comprising contig comprising a 797-bp fragment a 797-bp factor fragmentof neochrome. of neochrome. Thisin fragment plant This fragment evolution. GivenGiven that neochrome that may have playedneochrome a major role in pro- then wasmay thenextended was extendedt0 indudehave to almost include thaplayed almost entire the orf entire using ORF aa combination usingmajor a combination of roleof in pro PCR (see below) and additional read mapping and assembly. motingmoting the diversification the of ferns diversification under the CretaceousATertiaiy PCR (see ofbelow) andferns additional read under maPPin9 and assembly, the Cretaceous/Tertiary angiospermangiosperm canopy canopy(3, 7, 8), our (3, study 7, has8), important our study impli- „Lhas . ,. important. . „ impli r- e .J.c. i . Cloning of Neochrome, Phototropin, Cloning and of Neochrome,Phytochrome. Phototropin, To and verify PHytochrome. empirically To verify empirically cations for the macroevolutionary significance of plant-to- , .. . . , „ ' , J - cations „ e the presencefor the of the macroevolutionary hornwort photoreceptor the presence of the hornwort genes significancephotoreceptorfound in genesthe foundtran in the of tran plant-to plantPiant HGT. TiLj 1. scriptomes and to obtain intron/exon scriptomes information, and to obtain intron/exon we cloned information, the genes we cloned from the genes from genomic DNA from five hornwort species (SI Appendix, Table S3). In addition, Materials ■ Iand genomic Methods DNA from five hornwort species (SI Appendix, Table S3). In addition, Materials ana Methods neochrome sequences neochrome sequenceswere wereobtained obtained from from 25 fern 25 species fern by PCRspecies and by PCR and Mining Transcriptomes Transcriptomes and Whole-Genome and Whole-Genome Sequences forSequences Homologs for of Neochrome,Homologs cloning (SI of Appendix, Neochrome, Table cloningS3). Genomic (SI Appendix, DNA was extractedTable S3). using Genomic the DNA was extracted using the Phototropin, and and Phytochrome. Phytochrome. All but All one but of onethe 434of thetranscriptomes 434 transcriptomes Qiagen used DNAeasy usedPlant QiagenMini Kit DNAeasy (Qiagen). ThePlant gene Mini fragments Kit (Qiagen). were ampli The gene fragments were ampli were generated generated by theby Onethe ThousandOne Thousand Plants Project Plants (1 Project KP; www.onekp.com); (1KP; fied www.onekp.com); using Phusion DNA fiedpolymerase using (NewPhusion England DNA Biolabs) polymerase or Denville (New England Biolabs) or Denville these transcriptomes transcriptomes were were derived derived from froma diverse a diverseselection selectionof brown Choice algae,of brownTaq (Denville). algae, The Choice primers Taq and (Denville). detailed PCRThe conditions primers areand sum detailed PCR conditions are sum red algae,algae, green green algae, algae, bryophytes, bryophytes, lycophytes, lycophytes, ferns, andferns, seed andplantsmarized seed (SI in plants SI Appendix, (SI marized Tables S3in andSI Appendix, S4. The amplified Tables products S3 and wereS4. The amplified products were Appendix, Table Table S1 S1).). Details Details on RNAon RNA extraction, extraction, sequencing, sequencing, and cloned assembly and into assembly Promega pGEM-Tcloned (Promega)into Promega and sequenced. pGEM-T (Promega) and sequenced. for 1KP1KP can can be be found found in inJohnson Johnson et al.et (40).al. (40).Additionally, Additionally, a whole-plant a whole-plant normalized Illumina lllumina transcriptome transcriptome library library was constructedwas constructed and sequencedGenome-Walking and sequenced In HornwortGenome-Walking Phototropin In and Hornwort Neochrome. Phototropin To rule out the and Neochrome. To rule out the for PteridiumPteridium aquilinum aquilinum using using pooled pooled RNA RNA from from six sporophyte six sporophyte possibility tissues tissues that the possibility phototropin that gene foundthe phototropin in hornworts genemight foundbe a partial in hornworts might be a partial (young sporelingsporeling leaf, leaf, rhizome rhizome tip, tip, fiddlehead, fiddlehead, mature mature sterile sterile pinnae, neochrome, pinnae, and we and used neochrome, inverse PCR (50) we to used obtain inverse the flanking PCR (50)genomic to obtain region. the flanking genomic region, pinnaepinnae with with developing developing and mature and sporangia). mature The sporangia). Pteridium tran The Pteridium Genomic DNA tran- of N. Genomic aenigmaticus DNA was of digested N. aenigmaticus by apol (New wasEngland digested by apol (New England scriptomescriptome was was assembled assembled using defaultusing parametersdefault parameters in the Trinity in RNA-seq the Biolabs)Trinity and RNA-seq self-iigated Biolabs) using andT4 DNA self-ligated ligase (New using England T4 Biolabs).DNA ligase Then (New England Biolabs). Then pipelinepipeline version version r2012-01-25p1 r2012-01-25p1 (41). The (41).sequencing The sequencingreads were deposited reads innestedwere depositedPCRs were conductedin nested on PCRs the circularizedwere conducted DNA. The on ampliconsthe circularized were DNA. The amplicons were NationalNational Center Center for Biotechnologyfor Biotechnology Information Information (NCBI) Sequence (NCBI) Read Ar clonedSequence using Read Promega Ar- clonedpGEM-T using and sequenced. Promega To pGEM-T search for and the sequenced. genes To search for the genes chivechive (SRA) (SRA) under experimentunder experiment SRX423244. SRX423244. flanking flanking neochrome neochrome in inN. aenigmaticus,N. aenigmaticus, we used the Clontechwe used GenomeWalker the Clontech GenomeWalker ForFor the the 1 KP1KP transcriptomes, transcriptomes, we used weboth used Short bothOligonucleotide Short Oligonucleotide Assembly kit (Clontech) Assembly and followed kit (Clontech) the manufacturer's and followed manual. the The manufacturer's resulting PCR manual. The resulting PCR PackagePackage (SOAP) (SOAP) de novo de novoand SOAP and de SOAPnovo trans de novoassemblies. trans For assemblies. each as amplicons For eachwere clonedas- amplicons and sequenced. were In cloned total, we and obtained sequenced. 3,291-bp In andtotal, we obtained 3,291-bp and sembly,sembly, a BLASTa BLAST database database was constructed was constructed using the BLAST+ using packagethe BLAST+ (42). 4,578-bp package regions (42). up- 4,578-bp and down-stream, regions up-respectively, and down-stream, of neochrome. respectively, The of neochrome. The Neochrome, phototropin, phototropin, and phytochrome and phytochrome sequences weresequences queried weresepa primers queried for sepa-the above primers PCR reactions for the are above listed inPCR SI Appendix, reactions Table are S4.listed in SI Appendix, Table S4. rately (by (by tBLASTn tBLASTn for 1for KP and1 KP by andBLASTn by forBLASTn Pteridium for assemblies), Pteridium and assemblies), the and the significant hits hits to transcriptome to transcriptome scaffolds scaffolds were extracted were (e-value extracted threshold (e-value Sequence threshold Alignment Sequence for Neochrome, Alignment Phototropin, for Neochrome, and Phytochrome. Phototropin, We built and Phytochrome. We built ofof <1(T5). <10 5). For For each each scaffold, scaffold, the best the ORF bestwas identified, ORF was the identified, sequence was twothe largesequence alignments was two for phototropinlarge alignments and phytochrome, for phototropin with each and align phytochrome, with each align translated into into amino amino acids, acids,and then and BLASTp then queried BLASTp against queried the NCBI against ment including the NCBI the mentcorresponding including domains the fromcorresponding hornwort and domains fern neo from hornwort and fern neo nonredundantnonredundant protein protein database database (nr). The (nr). scaffolds The were scaffolds discarded were if they discarded chrome. The if phototropinthey chrome. dataset The contains phototropin 163 sequences dataset from contains 106 species, 163 sequences from 106 species, diddid not not match match neochrome, neochrome, phototropin, phototropin, or phytochrome or phytochromehomologs in the nr andhomologs the phytochrome in the nr dataset and the includes phytochrome 139 sequences dataset from includes76 species. 139 To sequences from 76 species. To databasedatabase with with an e-valuean e-value threshold threshold of <0.001. of For <0.001. 1KP transcriptomes, For 1KP transcriptomes, the reduce ambiguities the in reduce sequence ambiguities alignment, wein includedsequence only alignment, the conserved we included only the conserved filteredfiltered scaffolds scaffolds from SOAPfrom de novoSOAP and de SOAP novo de novoand trans SOAP assemblies de novo transdomains assemblies (i.e., LOV1, domainsLOV2, and (i.e., STK forLOV1, phototropins; LOV2, and PAS, STK GAF, for PHY, phototropins; PAS PAS, GAF, PHY, PAS were mergedmerged using using CAP3 CAP3 (43). (43). We carriedWe carried out the out above the proceduresabove procedures repeats, using HisKA, using and repeats, HATPase HisKA,for phytochromes). and HATPase The for domain phytochromes). boundaries The domain boundaries our PythonPython pipeline pipeline BlueDevil BlueDevil (see http://dx.doi.org/10.5061/dryad.fn2rg).(see http://dx.doi.org/10.5061/dryad.fn2rg). were identified by querying were identified each scaffold by against querying the NCBI each Conserved scaffold Doagainst the NCBI Conserved Do We alsoalso searched searched for forand obtainedand obtained photoreceptor photoreceptor homologs homologsfrom 39main plant fromDatabase 39 (51). plant Each main domain Database was aligned (51). separatelyEach domain (based was on alignedthe separately (based on the and algaealgae whole-genome whole-genome sequences sequences through through Phytozome Phytozome (44) and amino the (44) acid and sequences) the amino using acid Muscle sequences) (52) and then using was concatenated.Muscle (52) We and then was concatenated. We AmborellaAmborella Genome Genome Database Database (www.amborella.org) (www.amborella.org) (45). developed(45). developed a Python script, a Python DomainDivider script, (see DomainDivider http://dx.doi.org/10.5061/ (see http://dx.doi.org/10.5061/ dryad.fn2rg), to toautomate automate these these processes. processes. We also We generated also generated a separate a separate Assembling and and Mining Mining an Anthocerosan Anthoceros punctatus punctatus Draft Genome Draft Genomefor Homologs alignment for Homologs for hornwort alignment and fern for neochromes. hornwort This and alignment fern neochromes. was based on This alignment was based on of Neochrome,Neochrome, Phototropin, Phototropin, and Phytochrome. and Phytochrome. To generate To a generatedraft genome entire a draft for neochrome genome sequences for entire rather neochrome than on domains. sequences All alignmentsrather than were on domains. All alignments were Anthoceros punctatus, punctatus, genomic genomic DNA DNAwas sheared was sheared into ~400-bp into ~400-bpfragments inspected fragments manually, inspectedand ambiguously manually, aligned and regions ambiguously were excluded aligned before regions were excluded before and sequencedsequenced using using Illumina lllumina HiSeq2000, HiSeq2000, giving agiving total of a 25total million of phylogenetic 2590-bp million analyses. 90-bp Thephylogenetic phototropin, analyses. phytochrome, The phototropin,and neochrome phytochrome, and neochrome paired-end reads reads (about (about 20x 20xgenome genome coverage). coverage). The reads The were reads subjected alignments were subjected contain 1,716,alignments 2,802, and contain 4,002 bp,1,716, respectively. 2,802, andThe GenBank4,002 bp, respectively. The GenBank to twotwo cycles cycles of of read read error error correction correction using usingthe ALLPATHS-LG the ALLPATHS-LG FindError accession FindError numbers are accession listed in numbersSI Appendix, are Figs. listed S1-S3. in SI Appendix, Figs. S1-S3. program (46) (46) before before being being assembled assembled using using Velvet Velvet (47). Assemblies (47). Assemblies were were generated for for a rangea range of kmerof kmer values values (k = 21,(k 31,= 21, 41, 31, 51, 41,and 51,61) and Phylogenetic then61) and Analysesthen Phylogenetic of Phototropin Analyses and Neochrome. of Phototropin Phototropin and and neoNeochrome. Phototropin and neo were combined.combined. The The redundant redundant scaffolds scaffolds were removedwere removed using Usearch using chrome (48),Usearch phylogenies (48), werechrome inferred phylogenies based on their were nucleotide inferred alignments. based on We their nucleotide alignments. We and overlappingoverlapping contigs contigs were were subject subject to additional to additional assembly assembly using usedCAP3 using PartitionFinder to CAP3 to used(53) to PartitionFinder identify the optimal (53) data to partitionidentify schemes the optimal and data partition schemes and produce a adraft draft genome genome assembly. assembly. The finalThe assemblyfinal assembly contains contains 29,582 nucleotide con 29,582 substitution con- modelsnucleotide under the substitution Akaike Information models Criterion. under the Akaike Information Criterion, tigs withwith a atotal total combined combined assembly assembly length length of 99.5 ofMb 99.5 and theMb N50and Based contigthe on N50 this contig analysis, Based each on codon this position analysis, was each treated codon as a positiondistinct partiwas treated as a distinct parti length of of 4,955 4,955 bp. bp.The Thecontig contig length length ranges rangesfrom 919 from bp to 91976.5 bpkb, tion. to with 76.5 For kb,phototropin, with tion. the For first, phototropin, second, and thirdthe first, positions second, were andassigned third positions were assigned 1,643 contigs contigs over over 10 kb.10 kb.The Themedian median and mean and assembledmean assembled contig coverageGTR+T+I contig substitutioncoverage GTR+T+I models; forsubstitution neochrome, models; GTR+r+l, for GTR+T+I, neochrome, GTR+I GTR+r+l, GTR+T+I, GTR+I isis 18.1x18.1x and and 44x, 44x,respectively. respectively. The raw reads The were raw deposited reads werein NCBI deposited models in were NCBI applied models to each were codon applied position to respectively.each codon Weposition used Garli respectively. (54) We used Garli (54) underunder SRA096687. SRA096687. to obtain the maximum likelihood to obtain tree the maximumunder thelikelihood aforementioned tree under the aforementioned models. models, This assembly assembly was wassearched searched for homologs for homologs of neochrome, of neochrome, phototropin, with andphototropin, genthreshfortopoterm and with setgenthreshfortopoterm to 1,000,000 and eight independentset to 1,000,000 runs. and eight independent runs, phytochromephytochrome using using tBLASTn. tBLASTn. Although Although phototropin phototropin and phytochrome and Multiparametric genesphytochrome bootstrappinggenes Multiparametric was done using bootstrapping RAxML (55) withwas 1,000done using RAxML (55) with 1,000 were readily identified, no contig was found containing a putative neo o were readily identified, no contig was found containing replicates. a putative For the neo- neochrome replicates. alignment, For the we neochrome also carried outalignment, the tsame we also carried out the same 3 chrome sequence. sequence. To searchTo search for any for A. any punctatus A. punctatus neochrome neochrome gene maximum that gene likelihood that maximum analyses on likelihoodthe first + second-codonanalyses on positions the first —I and + on second-codon positions and on O perhaps failed failed to beto assembled, be assembled, all sequencing all sequencing reads were searchedreads were against the searchedthird-codon against positions the separately. third-codon We usedpositions MrBayes separately. (56) to conduct We used MrBayes (56) to conduct aa librarylibrary of ofneochrome neochrome protein protein sequences sequences using BLASTx. using Reads BLASTx. obtaining Bayesian Reads an obtaining tree inference an Bayesian under the tree same inference models, withunder two the independent same models, with two independent e-value of of <1(T10 <1(T10 were were isolated isolated and assembled and assembled using Velvet using with Velvet liberalMarkov with chain liberal Monte MarkovCarlo (MCMC) chain runs, Monte four Carlochains (MCMC)each, and runs,trees sam four chains each, and trees sam assembly parameters parameters (-cov_cutoff (-cov^cutoff 1 -min_pair_count 1 -min_pair_count 1 -edgeFractionCutoff 1 -edgeFractionCutoff pled every 1,000 generations. pled every Substitution 1,000 generations.parameters were Substitution unlinked, and parameters were unlinked, and 0.1 -scaffolding -scaffolding yes -min_contigJgthyes -min_contigJgth 90) at five 90) different at five values different for thekmer valuesrate prior for was kmer set tothe vary rate among prior partitions. was set to The vary MrBayes among output partitions. was The MrBayes output was length (21, (21, 31, 31, 41, 41,51, and51, 61). and The 61). resulting The resulting assemblies assemblieswere combined, inspected were combined, using Tracer inspected (57) to ensure using proper Tracer convergence (57) to ensure and mixing proper (ef convergence and mixing (ef redundant contigs contigs were werediscarded discarded using Usearch, using andUsearch, overlapping and contigs overlappingfective sample contigs sizes all fective>200), and sample 25% of sizesthe total all >200),generations and 25%were ofdis the total generations were dis were merged merged using using CAP3. CAP3. All sequencing All sequencing reads then reads were thenmapped were to carded these mapped as burn-in to these before carded making as burn-inthe 50% majority before makingconsensus the tree. 50% Because majority consensus tree. Because seed contigs contigs using using Bowtie2 Bowtie2 (49) with (49) thewith very-sensitive-local the very-sensitive-local option. thePaired stationary,option. Paired- homogeneous the stationary, assumptions homogeneous of GTR might beassumptions violated in casesof GTR might be violated in cases end reads reads where where at least at leastone read one mapped read mappedto the seed to contigs the seed were associated contigsse with were HGT andse- deep associated divergence with(58), we HGT also used and a nonstadeep divergence (58), we also used a nonsta lected.lected. All All the theselected selected reads then reads were then reassembled were reassembledas above. This mapping as tionary,above. heterogeneousThis mapping nucleotide tionary, substitution heterogeneous model implemented nucleotide in substitution model implemented in and assembly assembly process process was repeated was repeated until no untilfurther no reads further could bereads identified nhPhyML could be (59) identified to infer the nhPhyMLphototropin (59)tree. toThe infer analysis the was phototropincarried out tree. The analysis was carried out

Li Liet al. et al. PNAS | May 6, 2014 | vol.111 PNAS I| May no. 6,2014 |18 vol.111 | | no.6675 18 | 6675

This content downloaded from 152.3.102.254 on Mon, 30 Nov 2020 15:40:33 UTC All use subject to https://about.jstor.org/terms with 10 discrete categories of GC equilibrium frequencies, and the required Akaike Information Criterion. Maximum likelihood analyses were carried out starting tree was the best tree from the Garli analysis. To conduct boot in RAxML with 100 random starting trees, and multiparametric bootstrapping strapping in nhPhyML, we created a Python wrapper, and for each replicate, was done with 1,000 replicates. RAxML was used to input the starting tree. In addition to the nucleotide substitution model, we also used codon models to infer phylogenies, which Divergence Time Estimation of the Phototropin Gene Family. We used BEAST were carried out in CodonPhyML (19) under a maximum likelihood frame (69) to infer simultaneously the divergence times and phylogeny of the work. We used the Goldman-Yang (60) model with four categories of phototropin gene family. As recommended by PartitionFinder, the photo nonsynonymous/synonymous substitution rate ratios drawn from the dis tropin dataset was partitioned by codon position, each with the GTR+r+l crete gamma distribution, and codon frequencies were estimated from the substitution model. A total of 15 calibration priors were used (see SI Ap data under the F3 x 4 model (19). The tree topology search was done using pendix, Table S2 for details) (8, 13, 70-77), and a birth-death speciation prior the nearest neighbor interchange approach, and branch support was esti was used as the tree prior. We used the uncorrelated relaxed-dock model mated using the SH-like aLRT (61, 62) method. with rates drawn from a lognormal distribution. A starting tree was first estimated by r8s (78) and was provided to BEAST to initiate the run. Two Phylogenetic Analyses of Phytochrome. For the phytochrome phylogeny, we independent MCMC runs were carried out, and the output was inspected in used the protein alignment following the analytical strategy of Mathews Tracer to ensure convergence and mixing (effective sample sizes all >200). et al. (63). Using ProtTest (64), JTT + F was found to be the best empirical The trees from the two runs were combined in LogCombiner (69) with a 25% substitution model under the Akaike Information Criterion. For the maximum burn-in and were summarized in TreeAnnotator (69). It should be noted that likelihood analyses, we used Garli to search for the maximum likelihood tree, the stationary, homogeneous GTR model used here could be violated, es with genthreshfortopoterm set to 1,000,000 and eight independent runs, and pecially in the case of HGT, and might affect the divergence estimates. RAxML to conduct the multiparametric bootstrapping with 1,000 replicates. However, there is no nonstationary, heterogeneous model that is currently For Bayesian tree inference, we used MrBayes with two independent MCMC runs, four chains each, and trees sampled every 1,000 generations. After implemented in divergence time analyses, and our results should be revisited 25% of the total generations were removed, the 50% majority consensus in the future when more sophisticated methods are available. tree was calculated. Codon-based tree inference also was carried out as described above. Inferring Episodic Selection and GC Content Variation in Neochrome Evolution. To investigate whether fern neochromes had experienced pervasive episodic Topology Test. We used the Swofford-Olsen-Waddell-Hillis testpositive (65) selection,to we used the unrestricted, random effects branch-site compare the inferred HGT tree topology (i.e., fern neochromes model embedded (20) implemented in the HyPhy package (20, 79). Branches with epi within hornworts) against the alternative topologies suggestive sodic of vertical positive selection were identified by the sequential likelihood ratio test inheritance or independent origin, using the program sowhat (20). (66) Thewith neochrome alignment and the best maximum likelihood tree were RAxML and Seq-Gen (67). For testing the vertical inheritance topology,used as the to input data. The analyses were carried out on the Datamonkey pological constraints forcing fern and hornwort neochromes server to be (79, re 80). A GC content sliding window was constructed using a custom ciprocally monophyletic were used; for independent origin, constraints Python were script; each window is 400 bp in size, and the window slides every placed to have all fern genes to be monophyletic (i.e., monophyly 50 bp. either as neochrome + phototropin or neochrome + phytochrome). To calculate the posterior probability of the vertical transfer and independent ACKNOWLEDGMENTS.origin topol We thank Y.-L. Qiu, C.-W. Li, J. Nelson, B. Shaw, J. Duff, ogies, we filtered the posterior tree samples from MrBayes and D. Long,calculated and L. Forrest for helping us obtain hornwort materials; L. Huiet for the frequency of trees given the monophyly constraints. The filtering laboratory wasassistance; P. G. Wolf, C. W. dePamphilis, and P. E. Ralph for help in done by PAUP* (68). We applied this same approach to examine collecting the the pos Pteridium aquilinum transcriptome data; J. Leebens-Mack, terior distribution of fern neochrome gene trees. We searched S. Joya, for S. Ellis, to and others for contributing plant tissues for 1KP transcrip pologies that exhibited better congruence with the published tome species sequencing; D. L. Swofford and S. Wu for help with the phylogenetic analyses; and J. D. Palmer, M. Chen, J. Meireles, two anonymous reviewers, relationships (compared with the inferred gene tree). The constraint for members of the K.M.P. laboratory, and members of the Duke Systematics tree filtering required that neochromes from the same fern Discussionfamily Groupbe for insightful comments on a draft manuscript. This work monophyletic. was supported by research grants from the Society of Systematic Biologists, Sigma Xi, American Society of Plant Taxonomists, and Duke University (to Phylogenetic Analysis of the IGPD Gene. As a result of genome-walking F.-W.L.), National in Science Foundation Doctoral Dissertation Improvement N. aenigmaticus, we discovered an IGPD pseudogene downstream Grant DEB-1407158from (to F.-W.L. and K.M.P.), DEB-1145614 (to K.M.P.), and neochrome. To place this pseudogene in phylogenetic context, wea European resolved Research Council Advanced Investigator Award (to J.A.L.). The 1000 Plants (1KP) initiative, led by G.K.-S.W., is funded by the Alberta Min an IGPD phylogeny for land plants. A subset of the transcriptomes and istry of Enterprise and Advanced Education, Alberta Innovates Technology whole-genome sequences was mined for IGPD homologs (51 Appendix, Futures, InnovatesFig. Centre of Research Excellence, Musea Ventures, and BGI 54) using BlueDevil, and an alignment of IGPD (624 bp in length) Shenzhen. was conThis article is part of a doctoral dissertation in Biology at Duke structed manually. We partitioned the data by codon position, University with each by F.-W.L., supported by a National Science Foundation Grad partition given a GTR+F+I model as suggested by PartitionFinder uateunder Research the Fellowship.

1. Möglich A, Yang X, Ayers RA, Moffat K (2010) Structure and function of plant pho 10. Qiu Y-L, et al. (2006) The deepest divergences in land plants inferred from phyloge toreceptors. Annu Rev Plant Biol 61:21-47. nomic evidence. Proc Natl Acad Sei USA 103(42):15511-15516. 2. Nozue K, et al. (1998) A phytochrome from the fern Adiantum with features of the 11. Cox CJ, Li B, Foster PG, Embley TM, Civàn P (2014) Conflicting phylogenies for early putative photoreceptor NPH1. Proc Natl Acad Sei USA 95(26):15826-15830. land plants are caused by composition biases among synonymous substitutions. Syst 3. Kawai H, et al. (2003) Responses of ferns to red light are mediated by an un Biol 63(2):272-279. conventional photoreceptor. Nature 421 (6920):287-290. 12. Hedges SB, Kumar S (2009) The Timetree of Life (Oxford Univ Press, Oxford, UK). 4. Kanegae T, Hayashida E, Kuramoto C, Wada M (2006) A single chromoprotein with 13. Villarreal JC, Renner SS (2012) Hornwort pyrenoids, carbon-concentrating structures, triple chromophores acts as both a phytochrome and a phototropin. Proc Natl Acad evolved and were lost at least five times during the last 100 million years. Proc Natl Sei USA 103(47): 17997-18001. Acad Sei USA 109(46): 18873-18878. 5. Suetsugu N, Mittmann F, Wagner G, Hughes J, Wada M (2005) A chimeric photore 14. Schuettpelz E, Pryer KM (2007) Fern phylogeny inferred from 400 leptosporangiate ceptor gene, NEOCHROME, has arisen twice during plant evolution. Proc Natl Acad Sei species and three plastid genes. Taxon 56(4): 1037-1050. USA 102(38):13705-13709. 15. Rothfels CJ, et al. (2013) Transcriptome-mining for single-copy nuclear markers in 6. Yang Y, Qi X, Sen L, Su Y, Wang T (2010) Cloning and sequence analysis of red/blue ferns. PLoS ONE 8(10):e76957. light chimeric photoreceptor genes from three fern species (Coniogramme inter 16. Sanderson MJ, Shaffer HB (2002) Troubleshooting molecular phylogenetic analyses. media var. glabra, Plagiogyria distinctissima and Pronephrium lakhimpurnense). Am Annu Rev Ecol Syst 33:49-72. Fern J 100(1):1-15. 17. Kapralov MV, Filatov DA (2007) Widespread positive selection in the photosynthetic 7. Schneider H, et al. (2004) Ferns diversified in the shadow of angiosperms. Nature Rubisco enzyme. BMC Evol Biol 7:73. 428(6982):553-557. 18. Nabholz B, Künstner A, Wang R, Jarvis ED, Ellegren H (2011) Dynamic evolution of 8. Schuettpelz E, Pryer KM (2009) Evidence for a Cenozoic radiation of ferns in an an base composition: Causes and consequences in avian phylogenomics. Mol Biol Evol giosperm-dominated canopy. Proc Natl Acad Sei USA 106(27):11200-11205. 28(8):2197-2210. 9. Gontcharov AA, Melkonian M (2010) Molecular phytogeny and revision of the genus 19. Gil M, Zanetti MS, Zoller S, Anisimova M (2013) CodonPhyML: Fast maximum likeli Netrium (Zygnematophyceae, Streptophyta): Nucleotaenium gen. nov. J Phycol 46(2): hood phylogeny estimation under codon substitution models. Mol Biol Evol 30(6): 346-362. 1270-1280.

6676 j www.pnas.org/cgi/doi/10.1073/pnas.1319929111 Li et al.

This content downloaded from 152.3.102.254 on Mon, 30 Nov 2020 15:40:33 UTC All use subject to https://about.jstor.org/terms 20.20. Kosakovsky Kosakovsky Pond SL, et al. (2011) A random Pond effects branch-site SL, model et for detectingal. (2011) 50. Ochman 50. Ochman H, GerberA H, Gerber randomAS, Hartl AS, DL Hartl (1988) DL Genetic (1988)effects applicationsGenetic applications of anbranch-site inverse of anpolymerase inverse polymerase model for detecting episodicepisodic diversifying diversifying selection. Mol Biol Evol 28(11):3033-3043. selection. chain Molreaction. chain reaction. BiolGenetics Genetics 120(3):621-623.Evol 120(3):621-623. 28(11):3033-3043. 21.21. Lind PA,Lind Tobin C, BergPA, OG, KurlandTobin CG, Andersson C, DlBerg (2010) Compensatory OG, gene51.Kurland Marchler-Bauer 51. Marchler-Bauer A, etCG, al.A, (2011)et al. CDD:(2011)Andersson A CDD:conserved A conserved domain databasedomain DI fordatabase (2010) the functional for the functionalCompensatory gene amplificationamplification restores fitness after restores inter-species gene fitnessreplacements. Mol after Microbiol annotation annotation inter-species of proteins. of proteins. Nucleic Nucleic Acids ResAcids gene 39(Database Res 39(Database issue):D225-D229.replacements. issue):D225-D229. Mol Microbiol 75(5):1078-1089.75(5):1078-1089. 52. Edgar RC (2004) MUSCLE: Multiple 52. Edgarsequence RC (2004) alignment MUSCLE: Multiple with sequence high alignment accuracy with high and accuracy high and high 22. Näsvall Nasvall J, SunJ, SunL, Roth L, JR,Roth Andersson JR, Andersson Dl (2012) Real-time Dl (2012) evolution Real-time of new genesevolution by throughput.of new genes Nucleic by Acids throughput. Res 32(5):1792-1797. Nucleic Acids Res 32(5): 1792-1797. innovation,innovation, amplification, amplification, and divergence. and Sciencedivergence. 338(6105):384-387. Science 338(6105):384-387. 53. Lanfear 53. R, CalcottLanfear B, Ho R, SYW, Calcott Guindon B, SHo (2012) SYW, Partitionfinder: Guindon CombinedS (2012) selection Partitionfinder: Combined selection 23.23. Cho Cho Y, Y,Qiu QiuYL, KuhlmanYL, Kuhlman P, Palmer P, JD Palmer (1998) Explosive JD (1998) invasion Explosive of plant mitochondriainvasion of plant of partitioning mitochondria schemes of and partitioning substitution models schemes for phylogenetic and substitution analyses. Molmodels Biol for phylogenetic analyses. Mol Biol byby a groupa group I intron. I Proc intron. Natl Acad Proc Sei USA Natl 95(24): 14244-14249.Acad Sci USA 95(24): 14244-14249. Evol 29(6): 1695-1701. Evol 29(6):1695-1701. 24.24. Sanchez-Puerta Sanchez-Puerta MV, Cho MV, Y, MowerCho Y, JP, Mower Alverson JP, AJ, AlversonPalmer JD (2008) AJ, Frequent,Palmer JD 54. (2008) Zwickl Frequent, DJ (2006) Genetic 54. Zwickl Algorithm DJ (2006)Approaches Genetic for the Algorithm Phylogenetic Approaches Analysis of for the Phylogenetic Analysis of phylogeneticallyphylogenetically local horizontallocal horizontal transfer of transfer the coxl group of theI Intron coxl in groupflowering I Intron Large in Biologicalflowering Sequence Large Datasets Biological Under Sequence the Maximum Datasets Likelihood Under Criterion. the Maximum PhD Likelihood Criterion. PhD plantplant mitochondria. mitochondria. Mol Biol Evol 25(8):1762-1777. Mol Biol Evol 25(8):1762-1777. dissertation (Univ of Texas(Univ at Austin, of Texas Austin). at Austin, Austin). 25. Sanchez-Puerta Sanchez-Puerta MV, MV,et al. (2011)et al. Multiple(2011) Multiplerecent horizontal recent transfers horizontal of the transfers coxl 55. Stamatakis of the Acoxl (2006) 5$- RAxML-VI-HPC: Stamatakis MaximumA (2006) likelihood-basedRAxML-VI-HPC: phylogenetic Maximum analy likelihood-based phylogenetic analy intronintron in Solanaceaein Solanaceae and extended and extendedco-conversion co-conversion of flanking exons. of BMC flanking Evol Biol exons. ses with BMC thousands Evol Biolof taxa ses and thousands mixed models. of Bioinformatics taxa and mixed 22(21):2688-2690. models. Bioinformatics 22(21):2688-2690. 11:277.11:277. 56. Ronquist F, et al. (2012) MrBayes 3.2: 56.Efficient Ronquist F, et Bayesianal. (2012) MrBayes phylogenetic 3.2: Efficient Bayesian inference phylogenetic inferenceand and 26. JaedickeJaedicke K, K,Lichtenthäler Lichtenthaler AL, Meyberg AL, Meyberg R, Zeidler R, M, Zeidler Hughes JM, (2012) Hughes A phytochrome J model(2012) choice A phytochrome- across a large model model space. Systchoice Biol across61(3):539-542. a large model space. Syst Biol 61(3):539-542. 57. Rambaut A, Drummond AJ (2009) Tracer v1.5. Available at http://tree.bio.ed.ac.uk/ phototropin light light signaling signaling complex complex at the atplasma the membrane.plasma membrane. Proc Natl AcadProc Sei Natl USA Acad Sci USA ^7. Rambaut A, Drummond AJ (2009) Tracer v1.5. Available at http.//tree.bio.ed.ac.uk/ software/tracer/. Accessed June 2013. 109(30):12231-12236.109(30)12231-12236 software/tracer/. Accessed June 2013. 27.27. Burr Burr FA (1968) FA Chloroplast(1968) Chloroplast Structure and DivisionStructure in Megaceros and Species.Division PhD Disserin Megaceros 58.58. Verbyla Verbyla KL,Species. YapKL, VB, YapPhD Pahwa VB, Disser A, ShaoPahwa Y, Huttley A, Shao GA (2013) Y, Huttley The embedding GA problem(2013) forThe embedding problem for markov models of nucleotide substitution. PLoS ONE 8(7):e69187. tationtation (Univ of California,(Univ Berkeley). of California. Berkeley). markov models of nucleate substitution. PLoS ONE 8<7):e69187. 59.59. Boussau Boussau B, Gouy B, MGouy (2006) MEfficient (2006) likelihood Efficient computations likelihood with nonreversible computations with nonreversible 28.28. Huang Huang J (2013) J Horizontal(2013) Horizontal gene transfer ingene eukaryotes: transfer The weak-link in eukaryotes: model. Bio The weak-link model. Bio models of evolution. Syst Biol 55(5):756-768. essaysessays 35(10):868-875. 35(10) 868-875 models of evolution. Syst Biol 55(5):756-768. 60. GoldmanGoldman N, N,Yang Yang Z (1994) Z (1994) A codon-based A codon-based model of model nucleotide of nucleotide substitution substitution for for 29. Bergthorsson Bergthorsson U, Adams U, Adams KL, Thomason KL, Thomason B, Palmer JDB, (2003)Palmer Widespread JD (2003) horizontal Widespread horizontal protein-coding DNA sequences. Mol Biol Evol 11(5):725-736. transfertransfer of mitochondrialof mitochondrial genes in flowering genes plants.in flowering Nature 424(6945): plants. 197-201. Nature 424<6945):197-201. protein-coding DNA sequences. Mol Biol Evol 11<5):725-736. 61. Anisimova M, Gascuel O (2006) Approximate likelihood-ratio test for branches: A fast, 30.30. Bergthorsson Bergthorsson U, Richardson U, Richardson AO, Young AO, GJ, Young Goertzen GJ, LR, Goertzen Palmer JD LR, (2004) Palmer Massive JD (2004) Massive 61' Anisimova M, Ga«uel 0 <2006> Approximate likelihood-ratio test for branches: A fast. accurate, and and powerful powerful alternative. alternative. Syst Biol Syst 55(4):539-552. Biol 55(4):539-552. horizontal transfer transfer of mitochondrial of mitochondrial genes from genes diverse from land diverseplant donors land to plantthe donors to the 62. Guindon S, et al. (2010) New algorithms and methods to estimate maximum-likeli basalbasal angiosperm angiosperm Amborella. Amborella. Proc Natl AcadProc Sei Natl USA 101(51):17747-17752.Acad Sci USA 101 (51):17747-17752. 62' °uind°" et al' (2010) New algorithms and methods to estimate maximum-likeli hood phylogenies: Assessing the performance of PhyML 3.0. Syst Biol 59(3):307-321. 31.31. Davis Davis CC, CC, Wurdack Wurdack KJ (2004) KJ Host-to-parasite (2004) Host-to-parasite gene transfer genein flowering transfer plants: in CO flowering f 1 f. . ,r C plants:I « rv h 4 „r\ *ood D ft /IPhv'^emes: A /DA 1 At AAssessing rli m the performance of PhyML 3.0. Syst Biol 59(3):307-32T 63.63. Mathews Mathews S, Clements MD,S, BeilsteinClements MA (2010) AMD, duplicate Beilstein gene rooting of seedMA (2010) A duplicate gene rooting of seed PhylogeneticPhylogenetic evidence evidence from Malpighiales. from Science Malpighiales. 305(5684):676-678. Science 305(5684):676-678. plantsplants and the and phylogenetic the position phylogenetic of flowering plants. positionPhilos Trans R Socof Lond flowering B plants. Philos Trans R Soc Lond B 32.32. Davis Davis CC, AndersonCC, Anderson WR, Wurdack WR, KJ (2005) Wurdack Gene transfer KJ from(2005) a parasitic Gene flowering transfer from a parasitic flowering BiolBiol Sei 365(1539):383-395. Sci 365(1539):383-395. plantplant to ato fern. a fern.Proc Biol Proc Sei 272(1578):2237-2242. Biol Sci 272(1578):2237-2242. 64.64. Abascal Abascal F, Zardoya R,F, Posada Zardoya D (2005) ProtTest: R, SelectionPosada of best-fit D (2005) models of proteinProtTest: Selection of best-fit models of protein 33.33. Yoshida Yoshida S, Maruyama S, Maruyama S, Nozaki H, ShirasuS, Nozaki K (2010) H, Horizontal Shirasu gene Ktransfer (2010) by Horizontalthe gene transfer by the evolution.evolution. Bioinformatics Bioinformatics 21 (9):2104-2105. 21(9):2104-2105. parasiticparasitic plant plantStriga hermonthica. Striga hermonthica. Science 328(5982):1128. Science 328(5982):1128. 65.65. Goldman Goldman N, Anderson JP,N, Rodrigo Anderson AG (2000) Likelihood-based JP, Rodrigo tests of topologies AG in(2000) Likelihood-based tests of topologies in 34.34. Renner Renner SS, Bellot SS, S Bellot(2012) in GenomicsS (2012) of inChloroplasts Genomics and Mitochondria of Chloroplasts (Springer and Mitochondria (Springer phylogenetics.phylogenetics. Syst Biol 49(4):652-670. Syst Biol 49(4):652-670. Netherlands,Netherlands, Dordrecht), Dordrecht), Vol 35, Advances Vol in Photosynthesis 35, Advances and Respiration, in Photosynthesis and Respiration, 66.66. Church Church SH, Ryan JF, SH, Dunn CWRyan (2014) Sowhat.JF, AvailableDunn from CW GitHub (2014) repository Sowhat. Available from GitHub repository pp 223-235. , , . .. , , . n , https://github.com/josephryan/sowhat.AccessedJanuary2014. https://github.com/josephryan/sowhat. Accessed January 2014. 35. Xi Z, et al. (2013) Massive mitochondrial gene transfer in a parasitic flowering plant 35. X, Z, et al. (2013) Massive mitochondrial gene transfer 67. in Rambaut a parasitic A, Grassly flowering NC (1997) Seq-Gen: plant An 6? application Ramba(jt for the A MonteGrass|y Carlo Ncsimulation (1g97) Seq_Gen. An app|ication for the Monte Carlo simu|ation clade. PLoS Genet 9(2):e1003265. of DNADNA sequencesequence evolution evolution along along phylogenetic phylogenetic trees. trees. Comput Comput Appl Appl Biosci Biosci 13(3): 13(3): 36. Rice DW, et al. (2013) Horizontal transfer of entire genomes via mitochondrial fusion 235-238. in the angiosperm Amborella. Science 342(6165):1468-1473. 68. Swofford DLDL (2002)(2002) PAUP*. PAUP*. Phylogenetic Phylogenetic Analysis Analysis Using Using Parsimony Parsimony (*and (*and Other Other 37.37. Xi Xi Z, Z, et etal. (2012)al. (2012) Horizontal Horizontal transfer transfer of expressed of genes expressed in a parasitic genes flowering in a parasitic plant. flowering Methods). plant.Version Methods). 4.0al31 (SinauerVersion Associates,4.0al31 (Sinauer Sunderland, Associates, MA). Sunderland, MA). BMC Genomics 13:227. 69. DrummondDrummond AJ, AJ, Suchard Suchard MA, MA, Xie XieD, Rambaut D, Rambaut A (2012) A (2012) Bayesian Bayesian phylogenetics phylogenetics with with 38. Zhang Y,Y.etal. et al. (2013) (2013) Evolution Evolution of of a ahorizontally horizontally acquiredacquired legumelegume gene,gene, albuminalbumin BEAUti 1, in and BEAUti the BEAST and the 1.7. BEAST Mol Biol 1 7 EvolMol Biol29(8):1969-1973. Evol 29(8)1969-1973. the parasitic plantplant PhelipanchePhelipanche aegyptiacaaegyptiaca andand relatedrelated species.species. BMCBMC EvolEvol Biol70.Biol Clarke13:48. 13:48. ?0 JT, clarke Warnock JT Warnock RCM, DonoghueRCMi Donoghue PCJ (2011) PCj (2011) Establishing Establishing a time-scale a time-scale for forplant plant 39. Christin P-A,P-A, etet al.al. (2012)(2012) AdaptiveAdaptive evolutionevolution ofof C4C, photosynthesisphotosynthesis throughthrough recurrentevolution.recurrent evolution.New Phyto New I 192(1):266-301. Phytol 192(1):266-301. laterallateral gene gene transfer. transfer. Curr Biol Curr 22(5):445-449. Biol 22(5):445-449. 71. Hubers 71. M, Hubers Kerp M, H Kerp(2012) H (2012)Oldest Oldest known known mosses mosses discovered discovered inin Mississippian (late (late 40. Johnson MTJ, et al. (2012) Evaluating methodsmethods forfor isolatingisolating totaltotal RNARNA andand predictingpredictingVisean) strata Visean) of Germany.strata of Germany. Geo/ 40(8):755-758. Geo/ 40(8):755-758. the success ofof sequencingsequencing phylogeneticallyphylogenetically diverse diverse plant plant transcriptomes. transcriptomes. PLoS72. PLoS Guo ONE ONE C-Q, 72 Guo et al.C_Q (2012) et aL Riccardiothallus (2012) Riccardiothallus devonicus devonicus gen. et gen. sp. et nov., sp. nov.,the earliestthe earliest simple simple 7(11):e50226.7(11):e50226. thalloid liverwort from the Lower Devonian thalloid liverwortof Yunnan, from theChina. Lower DevonianRev Palaeobot of Yunnan, Palynol China. Rev Palaeobot Palynol 41. Grabherr Grabherr MG, MG, et al. et(2011) al. Full-length(2011) Full-length transcriptome transcriptome assembly from RNA-Seqassembly 176-177(C):35-40. data from RNA-Seq data 176-177(C)"35-40 withoutwithout a reference a reference genome. genome.Nat Biotechnol Nat 29(7):644-652. Biotechnol 29(7):644-652. 73. Prasad 73 V, prasadStrömberg V, CAE, Stromberg Alimohammadian CAE, AlimohammadianH, Sahni A (2005) Dinosaur H, Sahnicoprolites A (2005) Dinosaur coprolites 42. CamachoCamacho C, etC, al.et (2009) al. (2009) BLAST+: BLAST+: Architecture Architecture and applications. and applications. BMC Bioinformatics and BMC the earlyBioinformatics evolution of grassesand the and ear|y grazers. evolution Science of310(5751):1177—1180. grasses and grazers. Science 310(5751): 1177-1180. 10:421.10:421. 74. Kotyk ME, Basinger JF, Gensel PG, de 74. FreitasTA Kotyk ME, Basinger (2002) JF, Morphologically Gensel PG, de Freitas complexTA (2002) Morphologically plant complex plant 43.43. Huang Huang X, Madan X, Madan A (1999) ACAP3: (1999) A DNA CAP3: sequence A assemblyDNA sequence program. Genomeassembly Resmacrofossils program. from Genomethe Late Silurian Res macrofossils of Arctic Canada. from Am J theBot 89(6): Late 1004-1013. Silurian of Arctic Canada. Am J Bot 89(6): 1004-1013. 9(9):868-877.9(9):868-877. 75. skog JE, Banks HP (1973) Ibyka amphikoma,75. Skog JE, Banks gen. HP (1973)et sp. Ibyka n., amphikoma,a new protoarticulate gen. et sp. n., a new protoarticulate 44.44. Goodstein Goodstein DM, etDM, al. (2012) et al. Phytozome: (2012) Phytozome: A comparative platformA comparative for green plantplatform precursor fromfor thegreen late Middleplant Devonianprecursor of New from York the State. late Am Middle J Bot 60(4):366-380. Devonian of New York State. Am J Bot 60(4):366-380. genomics.genomics. Nucleic Nucleic Acids Res Acids40(Database Res issue):D1178-D1186. 40(Database issue):D1178-D1186. 76. Galtier J, Wang76. GaltierS-J, Li C-S, J, Hilton Wang J (2001) S-J, A newLi C-S,genus ofHilton filicalean J fern(2001) from A the new genus of filicalean fern from the 45.45. Amborella Amborella Genome Genome Project (2013)Project The (2013)Amborella The genome Amborella and the evolution genome ofLower and Permianthe evolution of China. Botof LowerJ Linn Soc Permian 137(4):429-442. of China. Bot J Linn Soc 137(4):429-442. flowering plants. plants. Science Science 342(6165):1241089. 342(6165):1241089. 77. Trivett ML (1992) Growth 77. Trivett architecture, ML (1992) Growthstructure, architecture, and relationships structure, of andCordaixylon relationships of Cordaixylon 46. MaccallumMaccallum I, etI, etal. (2009)al. (2009) ALLPATHS ALLPATHS 2: Small 2: genomesSmall genomes assembled assembled accurately accuratelyand with iowensis and nov. with comb. iowensis (Cordaitales). nov. comb. Int (Cordaitales).J Plant Sei 153(2):273-287. Int J Plant Sci 153(2):273-287. highhigh continuity continuity from from short pairedshort reads.paired Genome reads. Biol Genome 10(10):R103. Biol 10(10):R103. 78. Sanderson 78. Sanderson MJ (2003) MJ r8s:(2003) Inferring r8s: Inferring absolute absolute rates of ratesmolecular of molecular evolution evolutionand di and di 47. ZerbinoZerbino DR, DR, Birney Birney E (2008) E (2008) Velvet: Velvet: Algorithms Algorithms for de novo for shortde novo read shortassembly read using vergence assembly times using in vergence the absence times of ina molecular the absence clock. of aBioinformatics molecular clock. 19(2):301—302. Bioinformatics 19(2):301-302. de BruijnBruijn graphs. graphs. Genome Genome Res 18(5):821-829.Res 18(5):821-829. 79. Kosakovsky Pond 79.SL, KosakovskyFrost SDW, Pond Muse SL, SV Frost (2005) SDW, HyPhy: Muse Hypothesis SV (2005) HyPhy: testing Hypothesis using testing using 48. EdgarEdgar RC RC (2010) (2010) Search Search and and clustering clustering orders orders of magnitude of magnitude faster thanfaster BLAST. than BioBLAST. phylogenies. Bio- phylogenies. Bioinformatics Bioinformatics 21(5):676-679. 21(5):676-679. informaticsinformatics 26(19):2460-2461. 26(19):2460-2461. 80. Delport W, Poon AFY, Frost 80. Delport SDW, W, Kosakovsky Poon AFY, FrostPond SDW, SL (2010) Kosakovsky Datamonkey Pond SL (2010)2010: Datamonkey 2010: 49.49. Langmead Langmead B, Salzberg B, Salzberg SL (2012) SL Fast (2012) gapped-read Fast gapped-readalignment with alignmentBowtie 2. NatA suitewith of Bowtie phylogenetic 2. Nat analysis A suite tools offor phylogenetic evolutionary biology. analysis Bioinformatics tools for 26(19): evolutionary biology. Bioinformatics 26(19): MethodsMethods 9(4):357-359. 9(4):357-359. 2455-2457. 2455-2457.

Oo t— 3 -J-j Oo

Li Liet al. et al. PNAS | May 6, 2014 | vol.111 PNA5 I| Mayno. 6,2014 18| vol.111 | | 6677no. 18 | 6677

This content downloaded from 152.3.102.254 on Mon, 30 Nov 2020 15:40:33 UTC All use subject to https://about.jstor.org/terms