Dfzljdn9uc3pi.cloudfront.net



Supplementary Material forEvolution of digestive enzymes and dietary diversification in birdsYan-Hong Chen, and Huabin Zhao*Department of Ecology, Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, Wuhan University, Wuhan 430072, China*Corresponding Authors:Huabin ZhaoDepartment of Ecology, Hubei Key Laboratory of Cell Homeostasis,College of Life Sciences, Wuhan University299 Bayi Road, Wuhan, Hubei 430072, ChinaEmail address: HYPERLINK "mailto:huabinzhao@whu." huabinzhao@whu.Dietary dataFor a given species, the numerical value of each category represents the proportional usage of a given diet (Fig. 1 and Table S1). Given that most birds are general omnivores and the proportion of a particular component of food often varies significantly among species, we classified birds into two groups that contain species with relatively higher and lower consumption for one or multiple particular food components. Specifically, aiming to test whether the evolution of genes encoding proteases or lipases linked with meat consumption, birds were classified into two groups with contrasting meat consumption. In a similar fashion, aiming to investigate the evolution of amylases, species were divided into two groups with contrasting seed consumption; to investigate the evolution of carbohydrases except the amylases, species were divided into two groups with contrasting consumption of seeds, fruits or nectar; to investigate the evolution of enzymes involved in digesting insects, species were classified into two groups with contrasting insect consumption.Identification of digestive enzyme genesFirst, the complete protein sequences of each digestive enzyme gene were obtained from human, mouse, rat, alligator, chicken, turkey and zebra finch, and were used as queries for performing tblastn searches against each of 48 avian genomic sequences with the cutoff of e-value at 1e-10. Second, redundant sequences that hit the same genome regions were removed. Third, the retained blast hits mapping on the same scaffold were extended in both 5’ and 3’ directions and the exon/intron boundaries were predicted by the program GeneWise ADDIN EN.CITE <EndNote><Cite><Author>Birney</Author><Year>2004</Year><RecNum>199</RecNum><DisplayText>(Birney, Clamp &amp; Durbin 2004)</DisplayText><record><rec-number>199</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">199</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Birney, E.</author><author>Clamp, M.</author><author>Durbin, R.</author></authors></contributors><auth-address>European Bioinformat Inst, Cambridge CB10 1SA, England&#xD;Wellcome Trust Sanger Inst, Cambridge CB10 1SA, England</auth-address><titles><title>GeneWise and genomewise</title><secondary-title>Genome Res</secondary-title><alt-title>Genome Research</alt-title></titles><periodical><full-title>Genome Research</full-title><abbr-1>Genome Res</abbr-1></periodical><alt-periodical><full-title>Genome Research</full-title><abbr-1>Genome Res</abbr-1></alt-periodical><pages>988-995</pages><volume>14</volume><number>5</number><keywords><keyword>hidden markov model</keyword><keyword>DNA-sequences</keyword><keyword>mouse genome</keyword><keyword>prediction</keyword><keyword>alignment</keyword><keyword>drosophila</keyword><keyword>accuracy</keyword><keyword>profile</keyword></keywords><dates><year>2004</year><pub-dates><date>May</date></pub-dates></dates><isbn>1088-9051</isbn><accession-num>WOS:000221171700026</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000221171700026</url></related-urls></urls><electronic-resource-num>;(Birney, Clamp & Durbin 2004). Fourth, coding sequences shorter than 50% of the full length were discarded. Sequences covering the full sequences of query data with a putative start and stop codon were regarded as intact genes; those with partial coverage against the coding sequences from queries, were regarded as partial genes, which were characterized by an open reading frame (ORF); those with an interrupted reading frame were considered as pseudogenes. Fifth, the newly obtained genes were used as queries to conduct tblastn searches against the genomes, with the aim to identify additional genes. All candidate genes were ultimately verified by TBLASN searches against the entire GenBank database with their best blast hits as the known genes of targets.Phylogenetic analysisFor each type of digestive enzyme genes, protein sequences deduced from nucleotide sequences of all intact genes were first aligned with the program MUSCLE version 3.6 ADDIN EN.CITE <EndNote><Cite><Author>Edgar</Author><Year>2004</Year><RecNum>301</RecNum><DisplayText>(Edgar 2004)</DisplayText><record><rec-number>301</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">301</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Edgar, R. C.</author></authors></contributors><titles><title>MUSCLE: multiple sequence alignment with high accuracy and high throughput</title><secondary-title>Nucleic Acids Res</secondary-title><alt-title>Nucleic Acids Research</alt-title></titles><periodical><full-title>Nucleic Acids Research</full-title><abbr-1>Nucleic Acids Res</abbr-1></periodical><alt-periodical><full-title>Nucleic Acids Research</full-title><abbr-1>Nucleic Acids Res</abbr-1></alt-periodical><pages>1792-1797</pages><volume>32</volume><number>5</number><keywords><keyword>protein-structure</keyword><keyword>phylogenetic trees</keyword><keyword>psi-blast</keyword><keyword>database</keyword><keyword>programs</keyword><keyword>domains</keyword><keyword>smart</keyword><keyword>tool</keyword><keyword>identification</keyword><keyword>algorithms</keyword></keywords><dates><year>2004</year><pub-dates><date>Mar</date></pub-dates></dates><isbn>0305-1048</isbn><accession-num>WOS:000220487200025</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000220487200025</url></related-urls></urls><electronic-resource-num>;(Edgar 2004), and the alignments were subjected to manual inspection in MEGA version 6 ADDIN EN.CITE <EndNote><Cite><Author>Tamura</Author><Year>2013</Year><RecNum>302</RecNum><DisplayText>(Tamura et al. 2013)</DisplayText><record><rec-number>302</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">302</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Tamura, K.</author><author>Stecher, G.</author><author>Peterson, D.</author><author>Filipski, A.</author><author>Kumar, S.</author></authors></contributors><auth-address>Tokyo Metropolitan Univ, Res Ctr Genom &amp; Bioinformat, Hachioji, Tokyo, Japan&#xD;Tokyo Metropolitan Univ, Dept Biol Sci, Hachioji, Tokyo, Japan&#xD;Arizona State Univ, Biodesign Inst, Ctr Evolutionary Med &amp; Informat, Tempe, AZ 85287 USA&#xD;Arizona State Univ, Sch Life Sci, Tempe, AZ 85287 USA&#xD;King Abdulaziz Univ, Ctr Excellence Genom Med Res, Jeddah 21413, Saudi Arabia</auth-address><titles><title>MEGA6: molecular evolutionary genetics analysis version 6.0</title><secondary-title>Mol Biol Evol</secondary-title><alt-title>Molecular Biology and Evolution</alt-title></titles><periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></periodical><alt-periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></alt-periodical><pages>2725-2729</pages><volume>30</volume><number>12</number><keywords><keyword>software</keyword><keyword>relaxed clocks</keyword><keyword>phylogeny</keyword><keyword>divergence times</keyword></keywords><dates><year>2013</year><pub-dates><date>Dec</date></pub-dates></dates><isbn>0737-4038</isbn><accession-num>WOS:000327793000019</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000327793000019</url></related-urls></urls><electronic-resource-num>;(Tamura et al. 2013). Next we conducted codon alignment with PRANK version 130410 ADDIN EN.CITE <EndNote><Cite><Author>Loytynoja</Author><Year>2008</Year><RecNum>382</RecNum><DisplayText>(Loytynoja &amp; Goldman 2008)</DisplayText><record><rec-number>382</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">382</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Loytynoja, Ari</author><author>Goldman, Nick</author></authors></contributors><titles><title>Phylogeny-aware gap placement prevents errors in sequence alignment and evolutionary analysis</title><secondary-title>Science</secondary-title></titles><periodical><full-title>Science</full-title><abbr-1>Science</abbr-1></periodical><pages>1632-1635</pages><volume>320</volume><number>5883</number><dates><year>2008</year></dates><urls></urls></record></Cite></EndNote>(Loytynoja & Goldman 2008) and removed gaps from multiple sequence alignments using Gblocks version 0.91b ADDIN EN.CITE <EndNote><Cite><Author>Castresana</Author><Year>2000</Year><RecNum>1798</RecNum><DisplayText>(Castresana 2000)</DisplayText><record><rec-number>1798</rec-number><foreign-keys><key app="EN" db-id="ppz559pvwwtvtzefxe3vtxvsf59r9wd9deve">1798</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Castresana, J.</author></authors></contributors><auth-address>Castresana, J&#xD;European Mol Biol Lab, Meyerhofstr 1, D-69117 Heidelberg, Germany&#xD;European Mol Biol Lab, Meyerhofstr 1, D-69117 Heidelberg, Germany&#xD;European Mol Biol Lab, D-69117 Heidelberg, Germany</auth-address><titles><title>Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis</title><secondary-title>Molecular Biology and Evolution</secondary-title><alt-title>Mol Biol Evol</alt-title></titles><periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol. Biol. Evol.</abbr-1><abbr-2>Mol Biol Evol</abbr-2><abbr-3>Molecular Biology &amp; Evolution</abbr-3></periodical><alt-periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></alt-periodical><pages>540-552</pages><volume>17</volume><number>4</number><keywords><keyword>multiple alignments</keyword><keyword>conserved blocks</keyword><keyword>amino acid composition</keyword><keyword>mitochondrial proteins</keyword><keyword>eukaryotes</keyword><keyword>complete DNA-sequence</keyword><keyword>mitochondrial-DNA</keyword><keyword>nucleotide-sequence</keyword><keyword>genome organization</keyword><keyword>gene content</keyword><keyword>evolutionary link</keyword><keyword>fungi</keyword><keyword>proteins</keyword><keyword>metazoa</keyword><keyword>origins</keyword></keywords><dates><year>2000</year><pub-dates><date>Apr</date></pub-dates></dates><isbn>0737-4038</isbn><accession-num>ISI:000086337700008</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://000086337700008</url></related-urls></urls><language>English</language></record></Cite></EndNote>(Castresana 2000). Given that most partial chia genes cover the majority of intact sequences, partial chia genes span exon 4 to exon 10 (918 bp, adding to 64% of the total length), these partial genes were retained and combined with intact chia genes for phylogenetic reconstruction. Likewise, due to the difficulty to find out the complete sequences of one or multiple exons in other digestive enzyme genes, partial coding sequences spanning exon 2 to exon 33 of the agl (3,921 bp, adding to 85% of the total length), exon 3 to exon 19 of the gaa (1,800 bp, adding to 64% of the total length), and exon 2 to exon 8 of the gck (972 bp, adding to 70% of total length) were retained for subsequent analyses.Selective pressure analysesIn these analyses, the species tree derived from Prum et al. (2015) was used as the input tree for those single-copy genes, and the gene tree reconstructed by the Bayesian approach was used as the input tree for multi-copy chia or lyz (Fig. S1 and S2). In addition, evolutionary analyses of multi-copiy lyg and gaa were separately performed based on each subclade, namely lygA, lygB, lygC, gaa i and gaa ii, with the species tree applied as the input tree for each subclade (Fig. S3 and Fig. S4). Similarly, the pepsinogen C gene (pgc) was found to have two copies in most birds and was separately analyzed for each copy (we referred the nomenclature proposed by Castrol et al (2012) and renamed each copy as pgb1 and pgb2), with the species tree as the input tree for each copy (Fig. S5). Note that since all identified intact ctrc sequences were derived from species with no meat in their food composition (Fig. 1 and Table S1), we did not perform evolutionary analysis on this gene, which is required to classify species into two groups with relatively higher or lower meat consumption.When the two site-specific models (M8 and M8a) were compared, all sites under positive selection were identified using the Bayes Empirical Bayes (BEB) method with a cutoff of posterior probabilities ≥ 0.9 ADDIN EN.CITE <EndNote><Cite><Author>Yang</Author><Year>2007</Year><RecNum>203</RecNum><DisplayText>(Yang 2007)</DisplayText><record><rec-number>203</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">203</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Yang, Z. H.</author></authors></contributors><auth-address>UCL, Dept Biol, Galton Lab, London, England</auth-address><titles><title>PAML 4: Phylogenetic analysis by maximum likelihood</title><secondary-title>Mol Biol Evol</secondary-title><alt-title>Mol Biol Evol</alt-title></titles><periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></periodical><alt-periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></alt-periodical><pages>1586-1591</pages><volume>24</volume><number>8</number><keywords><keyword>codon models</keyword><keyword>likelihood</keyword><keyword>paml</keyword><keyword>phylogenetic analysis</keyword><keyword>software</keyword><keyword>amino-acid sites</keyword><keyword>detecting positive selection</keyword><keyword>codon-substitution models</keyword><keyword>episodic adaptive evolution</keyword><keyword>protein-coding sequences</keyword><keyword>molecular evolution</keyword><keyword>nucleotide substitution</keyword><keyword>divergence times</keyword><keyword>nonsynonymous substitution</keyword><keyword>statistical-methods</keyword></keywords><dates><year>2007</year><pub-dates><date>Aug</date></pub-dates></dates><isbn>0737-4038</isbn><accession-num>WOS:000248848400003</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000248848400003</url></related-urls></urls><electronic-resource-num>;(Yang 2007), and the likelihood ratio tests (LRT) between the two site models were conducted at a threshold of P<0.05. Simultaneously, three improved likelihood methods in the Datamonkey web server, with both nonsynonymous and synonymous substitution rates under consideration ADDIN EN.CITE <EndNote><Cite><Author>Pond</Author><Year>2005</Year><RecNum>204</RecNum><DisplayText>(Pond &amp; Frost 2005)</DisplayText><record><rec-number>204</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">204</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Pond, S. L. K.</author><author>Frost, S. D. W.</author></authors></contributors><auth-address>Univ Calif San Diego, Antiviral Res Ctr, San Diego, CA 92103 USA</auth-address><titles><title>Datamonkey: rapid detection of selective pressure on individual sites of codon alignments</title><secondary-title>Bioinformatics</secondary-title><alt-title>Bioinformatics</alt-title></titles><periodical><full-title>Bioinformatics</full-title><abbr-1>Bioinformatics</abbr-1></periodical><alt-periodical><full-title>Bioinformatics</full-title><abbr-1>Bioinformatics</abbr-1></alt-periodical><pages>2531-2533</pages><volume>21</volume><number>10</number><keywords><keyword>amino-acid sites</keyword><keyword>likelihood</keyword></keywords><dates><year>2005</year><pub-dates><date>May 15</date></pub-dates></dates><isbn>1367-4803</isbn><accession-num>WOS:000229285600054</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000229285600054</url></related-urls></urls><electronic-resource-num>;(Pond & Frost 2005), were used to further assess the signals of positive selection. Of the three methods, the fixed-effect likelihood (FEL) is a statistically rigorous method with both independent dN and dS inferred to each site in the context of codon substitution models ADDIN EN.CITE <EndNote><Cite><Author>Pond</Author><Year>2005</Year><RecNum>204</RecNum><DisplayText>(Pond &amp; Frost 2005)</DisplayText><record><rec-number>204</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">204</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Pond, S. L. K.</author><author>Frost, S. D. W.</author></authors></contributors><auth-address>Univ Calif San Diego, Antiviral Res Ctr, San Diego, CA 92103 USA</auth-address><titles><title>Datamonkey: rapid detection of selective pressure on individual sites of codon alignments</title><secondary-title>Bioinformatics</secondary-title><alt-title>Bioinformatics</alt-title></titles><periodical><full-title>Bioinformatics</full-title><abbr-1>Bioinformatics</abbr-1></periodical><alt-periodical><full-title>Bioinformatics</full-title><abbr-1>Bioinformatics</abbr-1></alt-periodical><pages>2531-2533</pages><volume>21</volume><number>10</number><keywords><keyword>amino-acid sites</keyword><keyword>likelihood</keyword></keywords><dates><year>2005</year><pub-dates><date>May 15</date></pub-dates></dates><isbn>1367-4803</isbn><accession-num>WOS:000229285600054</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000229285600054</url></related-urls></urls><electronic-resource-num>;(Pond & Frost 2005); the random-effect likelihood (REL) is an improvement of the Nielsen-Yang approach, which utilizes flexible but not overly parameter-rich rate distributions and permits both dS and dN to vary across codons independently ADDIN EN.CITE <EndNote><Cite><Author>Pond</Author><Year>2005</Year><RecNum>414</RecNum><DisplayText>(Pond &amp; Muse 2005)</DisplayText><record><rec-number>414</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">414</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Pond, Sergei L Kosakovsky</author><author>Muse, Spencer V</author></authors></contributors><titles><title>Site-to-site aariation of synonymous substitution rates</title><secondary-title>Molecular Biology and Evolution</secondary-title></titles><periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></periodical><pages>2375-2385</pages><volume>22</volume><number>12</number><dates><year>2005</year></dates><urls></urls></record></Cite></EndNote>(Pond & Muse 2005); the fast unconstrained Bayesian approximation (FUBAR), which can utilize several computational shortcuts to accelerate the detection of positive or purifying selection, and to relax the REL restrictions mentioned above, leads to improved robustness against model misspecification and allows analysis on large datasets ADDIN EN.CITE <EndNote><Cite><Author>Murrell</Author><Year>2013</Year><RecNum>413</RecNum><DisplayText>(Murrell et al. 2013)</DisplayText><record><rec-number>413</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">413</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Murrell, Ben</author><author>Moola, Sasha</author><author>Mabona, Amandla</author><author>Weighill, Thomas</author><author>Sheward, Daniel J</author><author>Pond, Sergei L Kosakovsky</author><author>Scheffler, Konrad</author></authors></contributors><titles><title>FUBAR: A fast, unconstrained bayesian approximation for inferring selection</title><secondary-title>Molecular Biology and Evolution</secondary-title></titles><periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></periodical><pages>1196-1205</pages><volume>30</volume><number>5</number><dates><year>2013</year></dates><urls></urls></record></Cite></EndNote>(Murrell et al. 2013). Sites with the significance level<0.2 for FEL, posterior probability>0.8 for FUBAR, or Bayes factor>50 for REL were chosen as candidates under positive selection. Finally, a complementary approach was adopted to identify changes in protein physicochemical properties using the TreeSAAP package ADDIN EN.CITE <EndNote><Cite><Author>Woolley</Author><Year>2003</Year><RecNum>205</RecNum><DisplayText>(Woolley et al. 2003)</DisplayText><record><rec-number>205</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">205</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Woolley, S.</author><author>Johnson, J.</author><author>Smith, M. J.</author><author>Crandall, K. A.</author><author>McClellan, D. A.</author></authors></contributors><auth-address>Brigham Young Univ, Dept Integrat Biol, Provo, UT 84602 USA&#xD;Brigham Young Univ, Dept Comp Sci, Provo, UT 84602 USA&#xD;Brigham Young Univ, Dept Microbiol &amp; Mol Biol, Provo, UT 84602 USA</auth-address><titles><title>TreeSAAP: selection on amino acid properties using phylogenetic trees</title><secondary-title>Bioinformatics</secondary-title><alt-title>Bioinformatics</alt-title></titles><periodical><full-title>Bioinformatics</full-title><abbr-1>Bioinformatics</abbr-1></periodical><alt-periodical><full-title>Bioinformatics</full-title><abbr-1>Bioinformatics</abbr-1></alt-periodical><pages>671-672</pages><volume>19</volume><number>5</number><keywords><keyword>codon-substitution models</keyword><keyword>protein evolution</keyword><keyword>sites</keyword><keyword>rates</keyword></keywords><dates><year>2003</year><pub-dates><date>Mar 22</date></pub-dates></dates><isbn>1367-4803</isbn><accession-num>WOS:000181964600024</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000181964600024</url></related-urls></urls><electronic-resource-num>;(Woolley et al. 2003), which performs goodness-of-fit and categorical statistical tests on 31 categories of structural and biochemical amino acid properties. For each gene, higher values (6-8) of each category of amino acid properties indicate more radical substitutions and more radical changes, which are signatures of adaptive evolution PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5TdW5hZ2FyPC9BdXRob3I+PFllYXI+MjAxMjwvWWVhcj48

UmVjTnVtPjIxMDwvUmVjTnVtPjxEaXNwbGF5VGV4dD4oU3VuYWdhciBldCBhbC4gMjAxMik8L0Rp

c3BsYXlUZXh0PjxyZWNvcmQ+PHJlYy1udW1iZXI+MjEwPC9yZWMtbnVtYmVyPjxmb3JlaWduLWtl

eXM+PGtleSBhcHA9IkVOIiBkYi1pZD0iMHN4c3BkZXJ0NXpwd2llOXowNnhwZmU3c2U5ZjBldGZk

ZGYwIj4yMTA8L2tleT48L2ZvcmVpZ24ta2V5cz48cmVmLXR5cGUgbmFtZT0iSm91cm5hbCBBcnRp

Y2xlIj4xNzwvcmVmLXR5cGU+PGNvbnRyaWJ1dG9ycz48YXV0aG9ycz48YXV0aG9yPlN1bmFnYXIs

IEsuPC9hdXRob3I+PGF1dGhvcj5Kb2huc29uLCBXLiBFLjwvYXV0aG9yPjxhdXRob3I+TyZhcG9z

O0JyaWVuLCBTLiBKLjwvYXV0aG9yPjxhdXRob3I+VmFzY29uY2Vsb3MsIFYuPC9hdXRob3I+PGF1

dGhvcj5BbnR1bmVzLCBBLjwvYXV0aG9yPjwvYXV0aG9ycz48L2NvbnRyaWJ1dG9ycz48YXV0aC1h

ZGRyZXNzPlVuaXYgUG9ydG8sIEN0ciBJbnRlcmRpc2NpcGxpbmFyIEludmVzdCBNYXJpbmhhICZh

bXA7IEFtYmllbnRhbCwgQ0lNQVIgQ0lJTUFSLCBQLTQxMDAgT3BvcnRvLCBQb3J0dWdhbCYjeEQ7

VW5pdiBQb3J0bywgRmFjIENpZW5jaWFzLCBEZXB0IEJpb2wsIFAtNDEwMCBPcG9ydG8sIFBvcnR1

Z2FsJiN4RDtOQ0ksIExhYiBHZW5vbSBEaXZlcnMsIEZyZWRlcmljaywgTUQgMjE3MDEgVVNBJiN4

RDtTdCBQZXRlcnNidXJnIFN0YXRlIFVuaXYsIFRoZW9kb3NpdXMgRG9iemhhbnNreSBDdHIgR2Vu

b21lIEJpb2luZm9ybWF0LCBTdCBQZXRlcnNidXJnLCBSdXNzaWE8L2F1dGgtYWRkcmVzcz48dGl0

bGVzPjx0aXRsZT5Fdm9sdXRpb24gb2YgQ1JJU1BzIGFzc29jaWF0ZWQgd2l0aCB0b3hpY29mZXJh

bi1yZXB0aWxpYW4gdmVub20gYW5kIG1hbW1hbGlhbiByZXByb2R1Y3Rpb248L3RpdGxlPjxzZWNv

bmRhcnktdGl0bGU+TW9sIEJpb2wgRXZvbDwvc2Vjb25kYXJ5LXRpdGxlPjxhbHQtdGl0bGU+TW9s

ZWN1bGFyIEJpb2xvZ3kgYW5kIEV2b2x1dGlvbjwvYWx0LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2Rp

Y2FsPjxmdWxsLXRpdGxlPk1vbGVjdWxhciBCaW9sb2d5IGFuZCBFdm9sdXRpb248L2Z1bGwtdGl0

bGU+PGFiYnItMT5Nb2wgQmlvbCBFdm9sPC9hYmJyLTE+PC9wZXJpb2RpY2FsPjxhbHQtcGVyaW9k

aWNhbD48ZnVsbC10aXRsZT5Nb2xlY3VsYXIgQmlvbG9neSBhbmQgRXZvbHV0aW9uPC9mdWxsLXRp

dGxlPjxhYmJyLTE+TW9sIEJpb2wgRXZvbDwvYWJici0xPjwvYWx0LXBlcmlvZGljYWw+PHBhZ2Vz

PjE4MDctMTgyMjwvcGFnZXM+PHZvbHVtZT4yOTwvdm9sdW1lPjxudW1iZXI+NzwvbnVtYmVyPjxr

ZXl3b3Jkcz48a2V5d29yZD5jcmlzcDwva2V5d29yZD48a2V5d29yZD5wb3NpdGl2ZSBzZWxlY3Rp

b248L2tleXdvcmQ+PGtleXdvcmQ+dG94aWNvZmVyYW4tcmVwdGlsaWFuIHZlbm9tIGV2b2x1dGlv

bjwva2V5d29yZD48a2V5d29yZD5hZGFwdGl2ZSBldm9sdXRpb248L2tleXdvcmQ+PGtleXdvcmQ+

Y3lzdGVpbmUtcmljaCBzZWNyZXRvcnkgcHJvdGVpbnM8L2tleXdvcmQ+PGtleXdvcmQ+bWFtbWFs

aWFuIHJlcHJvZHVjdGlvbjwva2V5d29yZD48a2V5d29yZD5tYW1tYWxpYW4gaW1tdW5lIHN5c3Rl

bTwva2V5d29yZD48a2V5d29yZD5yaWNoIHNlY3JldG9yeSBwcm90ZWluczwva2V5d29yZD48a2V5

d29yZD5hbWluby1hY2lkIHNpdGVzPC9rZXl3b3JkPjxrZXl3b3JkPmRldGVjdGluZyBtb2xlY3Vs

YXIgYWRhcHRhdGlvbjwva2V5d29yZD48a2V5d29yZD5jb2Rvbi1zdWJzdGl0dXRpb24gbW9kZWxz

PC9rZXl3b3JkPjxrZXl3b3JkPnNuYWtlLXZlbm9tPC9rZXl3b3JkPjxrZXl3b3JkPnBvc2l0aXZl

IHNlbGVjdGlvbjwva2V5d29yZD48a2V5d29yZD5waHlsb2dlbmV0aWMgYW5hbHlzaXM8L2tleXdv

cmQ+PGtleXdvcmQ+bWF4aW11bS1saWtlbGlob29kPC9rZXl3b3JkPjxrZXl3b3JkPmNyeXN0YWwt

c3RydWN0dXJlPC9rZXl3b3JkPjxrZXl3b3JkPmZ1bmN0aW9uYWwgZGl2ZXJnZW5jZTwva2V5d29y

ZD48L2tleXdvcmRzPjxkYXRlcz48eWVhcj4yMDEyPC95ZWFyPjxwdWItZGF0ZXM+PGRhdGU+SnVs

PC9kYXRlPjwvcHViLWRhdGVzPjwvZGF0ZXM+PGlzYm4+MDczNy00MDM4PC9pc2JuPjxhY2Nlc3Np

b24tbnVtPldPUzowMDAzMDU0MDkwMDAwMTI8L2FjY2Vzc2lvbi1udW0+PHVybHM+PHJlbGF0ZWQt

dXJscz48dXJsPiZsdDtHbyB0byBJU0kmZ3Q7Oi8vV09TOjAwMDMwNTQwOTAwMDAxMjwvdXJsPjwv

cmVsYXRlZC11cmxzPjwvdXJscz48ZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+aHR0cHM6Ly9kb2ku

b3JnLzEwLjEwOTMvbW9sYmV2L21zczA1ODwvZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+PGxhbmd1

YWdlPkVuZ2xpc2g8L2xhbmd1YWdlPjwvcmVjb3JkPjwvQ2l0ZT48L0VuZE5vdGU+

ADDIN EN.CITE PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5TdW5hZ2FyPC9BdXRob3I+PFllYXI+MjAxMjwvWWVhcj48

UmVjTnVtPjIxMDwvUmVjTnVtPjxEaXNwbGF5VGV4dD4oU3VuYWdhciBldCBhbC4gMjAxMik8L0Rp

c3BsYXlUZXh0PjxyZWNvcmQ+PHJlYy1udW1iZXI+MjEwPC9yZWMtbnVtYmVyPjxmb3JlaWduLWtl

eXM+PGtleSBhcHA9IkVOIiBkYi1pZD0iMHN4c3BkZXJ0NXpwd2llOXowNnhwZmU3c2U5ZjBldGZk

ZGYwIj4yMTA8L2tleT48L2ZvcmVpZ24ta2V5cz48cmVmLXR5cGUgbmFtZT0iSm91cm5hbCBBcnRp

Y2xlIj4xNzwvcmVmLXR5cGU+PGNvbnRyaWJ1dG9ycz48YXV0aG9ycz48YXV0aG9yPlN1bmFnYXIs

IEsuPC9hdXRob3I+PGF1dGhvcj5Kb2huc29uLCBXLiBFLjwvYXV0aG9yPjxhdXRob3I+TyZhcG9z

O0JyaWVuLCBTLiBKLjwvYXV0aG9yPjxhdXRob3I+VmFzY29uY2Vsb3MsIFYuPC9hdXRob3I+PGF1

dGhvcj5BbnR1bmVzLCBBLjwvYXV0aG9yPjwvYXV0aG9ycz48L2NvbnRyaWJ1dG9ycz48YXV0aC1h

ZGRyZXNzPlVuaXYgUG9ydG8sIEN0ciBJbnRlcmRpc2NpcGxpbmFyIEludmVzdCBNYXJpbmhhICZh

bXA7IEFtYmllbnRhbCwgQ0lNQVIgQ0lJTUFSLCBQLTQxMDAgT3BvcnRvLCBQb3J0dWdhbCYjeEQ7

VW5pdiBQb3J0bywgRmFjIENpZW5jaWFzLCBEZXB0IEJpb2wsIFAtNDEwMCBPcG9ydG8sIFBvcnR1

Z2FsJiN4RDtOQ0ksIExhYiBHZW5vbSBEaXZlcnMsIEZyZWRlcmljaywgTUQgMjE3MDEgVVNBJiN4

RDtTdCBQZXRlcnNidXJnIFN0YXRlIFVuaXYsIFRoZW9kb3NpdXMgRG9iemhhbnNreSBDdHIgR2Vu

b21lIEJpb2luZm9ybWF0LCBTdCBQZXRlcnNidXJnLCBSdXNzaWE8L2F1dGgtYWRkcmVzcz48dGl0

bGVzPjx0aXRsZT5Fdm9sdXRpb24gb2YgQ1JJU1BzIGFzc29jaWF0ZWQgd2l0aCB0b3hpY29mZXJh

bi1yZXB0aWxpYW4gdmVub20gYW5kIG1hbW1hbGlhbiByZXByb2R1Y3Rpb248L3RpdGxlPjxzZWNv

bmRhcnktdGl0bGU+TW9sIEJpb2wgRXZvbDwvc2Vjb25kYXJ5LXRpdGxlPjxhbHQtdGl0bGU+TW9s

ZWN1bGFyIEJpb2xvZ3kgYW5kIEV2b2x1dGlvbjwvYWx0LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2Rp

Y2FsPjxmdWxsLXRpdGxlPk1vbGVjdWxhciBCaW9sb2d5IGFuZCBFdm9sdXRpb248L2Z1bGwtdGl0

bGU+PGFiYnItMT5Nb2wgQmlvbCBFdm9sPC9hYmJyLTE+PC9wZXJpb2RpY2FsPjxhbHQtcGVyaW9k

aWNhbD48ZnVsbC10aXRsZT5Nb2xlY3VsYXIgQmlvbG9neSBhbmQgRXZvbHV0aW9uPC9mdWxsLXRp

dGxlPjxhYmJyLTE+TW9sIEJpb2wgRXZvbDwvYWJici0xPjwvYWx0LXBlcmlvZGljYWw+PHBhZ2Vz

PjE4MDctMTgyMjwvcGFnZXM+PHZvbHVtZT4yOTwvdm9sdW1lPjxudW1iZXI+NzwvbnVtYmVyPjxr

ZXl3b3Jkcz48a2V5d29yZD5jcmlzcDwva2V5d29yZD48a2V5d29yZD5wb3NpdGl2ZSBzZWxlY3Rp

b248L2tleXdvcmQ+PGtleXdvcmQ+dG94aWNvZmVyYW4tcmVwdGlsaWFuIHZlbm9tIGV2b2x1dGlv

bjwva2V5d29yZD48a2V5d29yZD5hZGFwdGl2ZSBldm9sdXRpb248L2tleXdvcmQ+PGtleXdvcmQ+

Y3lzdGVpbmUtcmljaCBzZWNyZXRvcnkgcHJvdGVpbnM8L2tleXdvcmQ+PGtleXdvcmQ+bWFtbWFs

aWFuIHJlcHJvZHVjdGlvbjwva2V5d29yZD48a2V5d29yZD5tYW1tYWxpYW4gaW1tdW5lIHN5c3Rl

bTwva2V5d29yZD48a2V5d29yZD5yaWNoIHNlY3JldG9yeSBwcm90ZWluczwva2V5d29yZD48a2V5

d29yZD5hbWluby1hY2lkIHNpdGVzPC9rZXl3b3JkPjxrZXl3b3JkPmRldGVjdGluZyBtb2xlY3Vs

YXIgYWRhcHRhdGlvbjwva2V5d29yZD48a2V5d29yZD5jb2Rvbi1zdWJzdGl0dXRpb24gbW9kZWxz

PC9rZXl3b3JkPjxrZXl3b3JkPnNuYWtlLXZlbm9tPC9rZXl3b3JkPjxrZXl3b3JkPnBvc2l0aXZl

IHNlbGVjdGlvbjwva2V5d29yZD48a2V5d29yZD5waHlsb2dlbmV0aWMgYW5hbHlzaXM8L2tleXdv

cmQ+PGtleXdvcmQ+bWF4aW11bS1saWtlbGlob29kPC9rZXl3b3JkPjxrZXl3b3JkPmNyeXN0YWwt

c3RydWN0dXJlPC9rZXl3b3JkPjxrZXl3b3JkPmZ1bmN0aW9uYWwgZGl2ZXJnZW5jZTwva2V5d29y

ZD48L2tleXdvcmRzPjxkYXRlcz48eWVhcj4yMDEyPC95ZWFyPjxwdWItZGF0ZXM+PGRhdGU+SnVs

PC9kYXRlPjwvcHViLWRhdGVzPjwvZGF0ZXM+PGlzYm4+MDczNy00MDM4PC9pc2JuPjxhY2Nlc3Np

b24tbnVtPldPUzowMDAzMDU0MDkwMDAwMTI8L2FjY2Vzc2lvbi1udW0+PHVybHM+PHJlbGF0ZWQt

dXJscz48dXJsPiZsdDtHbyB0byBJU0kmZ3Q7Oi8vV09TOjAwMDMwNTQwOTAwMDAxMjwvdXJsPjwv

cmVsYXRlZC11cmxzPjwvdXJscz48ZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+aHR0cHM6Ly9kb2ku

b3JnLzEwLjEwOTMvbW9sYmV2L21zczA1ODwvZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+PGxhbmd1

YWdlPkVuZ2xpc2g8L2xhbmd1YWdlPjwvcmVjb3JkPjwvQ2l0ZT48L0VuZE5vdGU+

ADDIN EN.CITE.DATA (Sunagar et al. 2012).The branch models performed by CODEML in the PAML package ADDIN EN.CITE <EndNote><Cite><Author>Yang</Author><Year>2007</Year><RecNum>203</RecNum><DisplayText>(Yang 2007)</DisplayText><record><rec-number>203</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">203</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Yang, Z. H.</author></authors></contributors><auth-address>UCL, Dept Biol, Galton Lab, London, England</auth-address><titles><title>PAML 4: Phylogenetic analysis by maximum likelihood</title><secondary-title>Mol Biol Evol</secondary-title><alt-title>Mol Biol Evol</alt-title></titles><periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></periodical><alt-periodical><full-title>Molecular Biology and Evolution</full-title><abbr-1>Mol Biol Evol</abbr-1></alt-periodical><pages>1586-1591</pages><volume>24</volume><number>8</number><keywords><keyword>codon models</keyword><keyword>likelihood</keyword><keyword>paml</keyword><keyword>phylogenetic analysis</keyword><keyword>software</keyword><keyword>amino-acid sites</keyword><keyword>detecting positive selection</keyword><keyword>codon-substitution models</keyword><keyword>episodic adaptive evolution</keyword><keyword>protein-coding sequences</keyword><keyword>molecular evolution</keyword><keyword>nucleotide substitution</keyword><keyword>divergence times</keyword><keyword>nonsynonymous substitution</keyword><keyword>statistical-methods</keyword></keywords><dates><year>2007</year><pub-dates><date>Aug</date></pub-dates></dates><isbn>0737-4038</isbn><accession-num>WOS:000248848400003</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000248848400003</url></related-urls></urls><electronic-resource-num>;(Yang 2007) were used to estimate ω values among branches. Due to inability to determine the dietary composition of ancestral lineages, the alternative model assigned an independent ω value to each ancestral branch. By contrast, the null model assumed that all branches possess a same ω value. The chi-square test was applied to examine the difference between the alternative and null models at a significant value of 0.05, using the false discovery rate (FDR) approach to adjust multiple testing ADDIN EN.CITE <EndNote><Cite><Author>Benjamini</Author><Year>1995</Year><RecNum>209</RecNum><DisplayText>(Benjamini &amp; Hochberg 1995)</DisplayText><record><rec-number>209</rec-number><foreign-keys><key app="EN" db-id="0sxspdert5zpwie9z06xpfe7se9f0etfddf0">209</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Benjamini, Y.</author><author>Hochberg, Y.</author></authors></contributors><titles><title>Controlling the false discovery rate - a practical and powerful approach to multiple testing</title><secondary-title>J Roy Stat Soc B Met</secondary-title><alt-title>Journal of the Royal Statistical Society Series B-Methodological</alt-title></titles><periodical><full-title>Journal of the Royal Statistical Society Series B-Methodological</full-title><abbr-1>J Roy Stat Soc B Met</abbr-1></periodical><alt-periodical><full-title>Journal of the Royal Statistical Society Series B-Methodological</full-title><abbr-1>J Roy Stat Soc B Met</abbr-1></alt-periodical><pages>289-300</pages><volume>57</volume><number>1</number><keywords><keyword>bonferroni-type procedures</keyword><keyword>familywise error rate</keyword><keyword>multiple-comparison procedures</keyword><keyword>p-values</keyword><keyword>bonferroni procedure</keyword></keywords><dates><year>1995</year></dates><isbn>0035-9246</isbn><accession-num>WOS:A1995QE45300017</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:A1995QE45300017</url></related-urls></urls><language>English</language></record></Cite></EndNote>(Benjamini & Hochberg 1995).The RELAX method (Wertheim et al. 2015) in the HyPhy package (Pond, Frost & Muse 2005) was used to test whether natural selection is intensified or relaxed along any of the branches in avian phylogeny. For each gene, we labeled branches with relatively lower consumption of particular food items as “test branches” and other branches were labeled as “reference branches”. The alternative model estimates the ω distribution for test branches or reference branches with the selection intensity parameter k, while the null model fixes the distribution of ω to be the same between the two categories of branches with k constrained to 1 PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5XZXJ0aGVpbTwvQXV0aG9yPjxZZWFyPjIwMTU8L1llYXI+

PFJlY051bT4yMDc8L1JlY051bT48RGlzcGxheVRleHQ+KFdlcnRoZWltIGV0IGFsLiAyMDE1KTwv

RGlzcGxheVRleHQ+PHJlY29yZD48cmVjLW51bWJlcj4yMDc8L3JlYy1udW1iZXI+PGZvcmVpZ24t

a2V5cz48a2V5IGFwcD0iRU4iIGRiLWlkPSIwc3hzcGRlcnQ1enB3aWU5ejA2eHBmZTdzZTlmMGV0

ZmRkZjAiPjIwNzwva2V5PjwvZm9yZWlnbi1rZXlzPjxyZWYtdHlwZSBuYW1lPSJKb3VybmFsIEFy

dGljbGUiPjE3PC9yZWYtdHlwZT48Y29udHJpYnV0b3JzPjxhdXRob3JzPjxhdXRob3I+V2VydGhl

aW0sIEouIE8uPC9hdXRob3I+PGF1dGhvcj5NdXJyZWxsLCBCLjwvYXV0aG9yPjxhdXRob3I+U21p

dGgsIE0uIEQuPC9hdXRob3I+PGF1dGhvcj5Qb25kLCBTLiBMLiBLLjwvYXV0aG9yPjxhdXRob3I+

U2NoZWZmbGVyLCBLLjwvYXV0aG9yPjwvYXV0aG9ycz48L2NvbnRyaWJ1dG9ycz48YXV0aC1hZGRy

ZXNzPlVuaXYgQ2FsaWYgU2FuIERpZWdvLCBEZXB0IE1lZCwgU2FuIERpZWdvLCBDQSA5MjEwMyBV

U0EmI3hEO1VuaXYgQ2FsaWYgU2FuIERpZWdvLCBCaW9pbmZvcm1hdCAmYW1wOyBTeXN0IEJpb2wg

R3JhZCBQcm9ncmFtLCBTYW4gRGllZ28sIENBIDkyMTAzIFVTQSYjeEQ7VW5pdiBTdGVsbGVuYm9z

Y2gsIERlcHQgTWF0aCBTY2ksIFpBLTc2MDAgU3RlbGxlbmJvc2NoLCBTb3V0aCBBZnJpY2E8L2F1

dGgtYWRkcmVzcz48dGl0bGVzPjx0aXRsZT5SRUxBWDogZGV0ZWN0aW5nIHJlbGF4ZWQgc2VsZWN0

aW9uIGluIGEgcGh5bG9nZW5ldGljIGZyYW1ld29yazwvdGl0bGU+PHNlY29uZGFyeS10aXRsZT5N

b2wgQmlvbCBFdm9sPC9zZWNvbmRhcnktdGl0bGU+PGFsdC10aXRsZT5Nb2xlY3VsYXIgQmlvbG9n

eSBhbmQgRXZvbHV0aW9uPC9hbHQtdGl0bGU+PC90aXRsZXM+PHBlcmlvZGljYWw+PGZ1bGwtdGl0

bGU+TW9sZWN1bGFyIEJpb2xvZ3kgYW5kIEV2b2x1dGlvbjwvZnVsbC10aXRsZT48YWJici0xPk1v

bCBCaW9sIEV2b2w8L2FiYnItMT48L3BlcmlvZGljYWw+PGFsdC1wZXJpb2RpY2FsPjxmdWxsLXRp

dGxlPk1vbGVjdWxhciBCaW9sb2d5IGFuZCBFdm9sdXRpb248L2Z1bGwtdGl0bGU+PGFiYnItMT5N

b2wgQmlvbCBFdm9sPC9hYmJyLTE+PC9hbHQtcGVyaW9kaWNhbD48cGFnZXM+ODIwLTgzMjwvcGFn

ZXM+PHZvbHVtZT4zMjwvdm9sdW1lPjxudW1iZXI+MzwvbnVtYmVyPjxrZXl3b3Jkcz48a2V5d29y

ZD5yZWxheGVkIHNlbGVjdGlvbjwva2V5d29yZD48a2V5d29yZD5tb2xlY3VsYXIgZXZvbHV0aW9u

PC9rZXl3b3JkPjxrZXl3b3JkPmNvZG9uIG1vZGVsczwva2V5d29yZD48a2V5d29yZD5waHlsb2dl

bmV0aWNzPC9rZXl3b3JkPjxrZXl3b3JkPmVuZG9zeW1iaW9udHM8L2tleXdvcmQ+PGtleXdvcmQ+

ZWNob2xvY2F0aW9uPC9rZXl3b3JkPjxrZXl3b3JkPm9wc2luPC9rZXl3b3JkPjxrZXl3b3JkPmRh

cGhuaWEgcHVsZXg8L2tleXdvcmQ+PGtleXdvcmQ+Ym9ybmF2aXJ1czwva2V5d29yZD48a2V5d29y

ZD5jb2Rvbi1zdWJzdGl0dXRpb24gbW9kZWxzPC9rZXl3b3JkPjxrZXl3b3JkPnRhc3RlIHJlY2Vw

dG9yIGdlbmVzPC9rZXl3b3JkPjxrZXl3b3JkPmFtaW5vLWFjaWQgc2l0ZXM8L2tleXdvcmQ+PGtl

eXdvcmQ+cHVyaWZ5aW5nIHNlbGVjdGlvbjwva2V5d29yZD48a2V5d29yZD5ETkEtc2VxdWVuY2Vz

PC9rZXl3b3JkPjxrZXl3b3JkPmNvbG9yLXZpc2lvbjwva2V5d29yZD48a2V5d29yZD5udWNsZW90

aWRlIHN1YnN0aXR1dGlvbjwva2V5d29yZD48a2V5d29yZD5iYWN0ZXJpYWwgZW5kb3N5bWJpb250

czwva2V5d29yZD48a2V5d29yZD5mYWN1bHRhdGl2ZSBzeW1iaW9udHM8L2tleXdvcmQ+PGtleXdv

cmQ+cGhlbm90eXBpYyBwbGFzdGljaXR5PC9rZXl3b3JkPjwva2V5d29yZHM+PGRhdGVzPjx5ZWFy

PjIwMTU8L3llYXI+PHB1Yi1kYXRlcz48ZGF0ZT5NYXI8L2RhdGU+PC9wdWItZGF0ZXM+PC9kYXRl

cz48aXNibj4wNzM3LTQwMzg8L2lzYm4+PGFjY2Vzc2lvbi1udW0+V09TOjAwMDM1MDkwODUwMDAy

MTwvYWNjZXNzaW9uLW51bT48dXJscz48cmVsYXRlZC11cmxzPjx1cmw+Jmx0O0dvIHRvIElTSSZn

dDs6Ly9XT1M6MDAwMzUwOTA4NTAwMDIxPC91cmw+PC9yZWxhdGVkLXVybHM+PC91cmxzPjxlbGVj

dHJvbmljLXJlc291cmNlLW51bT5odHRwczovL2RvaS5vcmcvMTAuMTA5My9tb2xiZXYvbXN1NDAw

PC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48bGFuZ3VhZ2U+RW5nbGlzaDwvbGFuZ3VhZ2U+PC9y

ZWNvcmQ+PC9DaXRlPjwvRW5kTm90ZT4A

ADDIN EN.CITE PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5XZXJ0aGVpbTwvQXV0aG9yPjxZZWFyPjIwMTU8L1llYXI+

PFJlY051bT4yMDc8L1JlY051bT48RGlzcGxheVRleHQ+KFdlcnRoZWltIGV0IGFsLiAyMDE1KTwv

RGlzcGxheVRleHQ+PHJlY29yZD48cmVjLW51bWJlcj4yMDc8L3JlYy1udW1iZXI+PGZvcmVpZ24t

a2V5cz48a2V5IGFwcD0iRU4iIGRiLWlkPSIwc3hzcGRlcnQ1enB3aWU5ejA2eHBmZTdzZTlmMGV0

ZmRkZjAiPjIwNzwva2V5PjwvZm9yZWlnbi1rZXlzPjxyZWYtdHlwZSBuYW1lPSJKb3VybmFsIEFy

dGljbGUiPjE3PC9yZWYtdHlwZT48Y29udHJpYnV0b3JzPjxhdXRob3JzPjxhdXRob3I+V2VydGhl

aW0sIEouIE8uPC9hdXRob3I+PGF1dGhvcj5NdXJyZWxsLCBCLjwvYXV0aG9yPjxhdXRob3I+U21p

dGgsIE0uIEQuPC9hdXRob3I+PGF1dGhvcj5Qb25kLCBTLiBMLiBLLjwvYXV0aG9yPjxhdXRob3I+

U2NoZWZmbGVyLCBLLjwvYXV0aG9yPjwvYXV0aG9ycz48L2NvbnRyaWJ1dG9ycz48YXV0aC1hZGRy

ZXNzPlVuaXYgQ2FsaWYgU2FuIERpZWdvLCBEZXB0IE1lZCwgU2FuIERpZWdvLCBDQSA5MjEwMyBV

U0EmI3hEO1VuaXYgQ2FsaWYgU2FuIERpZWdvLCBCaW9pbmZvcm1hdCAmYW1wOyBTeXN0IEJpb2wg

R3JhZCBQcm9ncmFtLCBTYW4gRGllZ28sIENBIDkyMTAzIFVTQSYjeEQ7VW5pdiBTdGVsbGVuYm9z

Y2gsIERlcHQgTWF0aCBTY2ksIFpBLTc2MDAgU3RlbGxlbmJvc2NoLCBTb3V0aCBBZnJpY2E8L2F1

dGgtYWRkcmVzcz48dGl0bGVzPjx0aXRsZT5SRUxBWDogZGV0ZWN0aW5nIHJlbGF4ZWQgc2VsZWN0

aW9uIGluIGEgcGh5bG9nZW5ldGljIGZyYW1ld29yazwvdGl0bGU+PHNlY29uZGFyeS10aXRsZT5N

b2wgQmlvbCBFdm9sPC9zZWNvbmRhcnktdGl0bGU+PGFsdC10aXRsZT5Nb2xlY3VsYXIgQmlvbG9n

eSBhbmQgRXZvbHV0aW9uPC9hbHQtdGl0bGU+PC90aXRsZXM+PHBlcmlvZGljYWw+PGZ1bGwtdGl0

bGU+TW9sZWN1bGFyIEJpb2xvZ3kgYW5kIEV2b2x1dGlvbjwvZnVsbC10aXRsZT48YWJici0xPk1v

bCBCaW9sIEV2b2w8L2FiYnItMT48L3BlcmlvZGljYWw+PGFsdC1wZXJpb2RpY2FsPjxmdWxsLXRp

dGxlPk1vbGVjdWxhciBCaW9sb2d5IGFuZCBFdm9sdXRpb248L2Z1bGwtdGl0bGU+PGFiYnItMT5N

b2wgQmlvbCBFdm9sPC9hYmJyLTE+PC9hbHQtcGVyaW9kaWNhbD48cGFnZXM+ODIwLTgzMjwvcGFn

ZXM+PHZvbHVtZT4zMjwvdm9sdW1lPjxudW1iZXI+MzwvbnVtYmVyPjxrZXl3b3Jkcz48a2V5d29y

ZD5yZWxheGVkIHNlbGVjdGlvbjwva2V5d29yZD48a2V5d29yZD5tb2xlY3VsYXIgZXZvbHV0aW9u

PC9rZXl3b3JkPjxrZXl3b3JkPmNvZG9uIG1vZGVsczwva2V5d29yZD48a2V5d29yZD5waHlsb2dl

bmV0aWNzPC9rZXl3b3JkPjxrZXl3b3JkPmVuZG9zeW1iaW9udHM8L2tleXdvcmQ+PGtleXdvcmQ+

ZWNob2xvY2F0aW9uPC9rZXl3b3JkPjxrZXl3b3JkPm9wc2luPC9rZXl3b3JkPjxrZXl3b3JkPmRh

cGhuaWEgcHVsZXg8L2tleXdvcmQ+PGtleXdvcmQ+Ym9ybmF2aXJ1czwva2V5d29yZD48a2V5d29y

ZD5jb2Rvbi1zdWJzdGl0dXRpb24gbW9kZWxzPC9rZXl3b3JkPjxrZXl3b3JkPnRhc3RlIHJlY2Vw

dG9yIGdlbmVzPC9rZXl3b3JkPjxrZXl3b3JkPmFtaW5vLWFjaWQgc2l0ZXM8L2tleXdvcmQ+PGtl

eXdvcmQ+cHVyaWZ5aW5nIHNlbGVjdGlvbjwva2V5d29yZD48a2V5d29yZD5ETkEtc2VxdWVuY2Vz

PC9rZXl3b3JkPjxrZXl3b3JkPmNvbG9yLXZpc2lvbjwva2V5d29yZD48a2V5d29yZD5udWNsZW90

aWRlIHN1YnN0aXR1dGlvbjwva2V5d29yZD48a2V5d29yZD5iYWN0ZXJpYWwgZW5kb3N5bWJpb250

czwva2V5d29yZD48a2V5d29yZD5mYWN1bHRhdGl2ZSBzeW1iaW9udHM8L2tleXdvcmQ+PGtleXdv

cmQ+cGhlbm90eXBpYyBwbGFzdGljaXR5PC9rZXl3b3JkPjwva2V5d29yZHM+PGRhdGVzPjx5ZWFy

PjIwMTU8L3llYXI+PHB1Yi1kYXRlcz48ZGF0ZT5NYXI8L2RhdGU+PC9wdWItZGF0ZXM+PC9kYXRl

cz48aXNibj4wNzM3LTQwMzg8L2lzYm4+PGFjY2Vzc2lvbi1udW0+V09TOjAwMDM1MDkwODUwMDAy

MTwvYWNjZXNzaW9uLW51bT48dXJscz48cmVsYXRlZC11cmxzPjx1cmw+Jmx0O0dvIHRvIElTSSZn

dDs6Ly9XT1M6MDAwMzUwOTA4NTAwMDIxPC91cmw+PC9yZWxhdGVkLXVybHM+PC91cmxzPjxlbGVj

dHJvbmljLXJlc291cmNlLW51bT5odHRwczovL2RvaS5vcmcvMTAuMTA5My9tb2xiZXYvbXN1NDAw

PC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48bGFuZ3VhZ2U+RW5nbGlzaDwvbGFuZ3VhZ2U+PC9y

ZWNvcmQ+PC9DaXRlPjwvRW5kTm90ZT4A

ADDIN EN.CITE.DATA (Wertheim et al. 2015). If the P value is less than 0.05 (calculated by a LRT), it implies that natural selection on the test branches has been relaxed (k < 1) or intensified (k > 1) relative to the reference branches. Furthermore, the less constrained partitioned exploratory model, which permits the proportion of sites and ω from each category of branches to change regardless of the selection intensity parameter (k), was used to estimate the ω distribution for the two categories of branches. Intensified selection was inferred as ω away from neutrality (ω = 1), while relaxed selection was predicted as ω converging to neutrality (ω = 1).Functional divergence testingThe coefficients of functional divergence (θI and θII) represent the levels of type-I and type-II divergence, respectively. Sites with a posterior probability (Qk)>0.95 were selected as functional divergence-related candidates. Chi-square distribution and normal distribution were applied to estimate the likelihood ratio of type-I analysis and the z-score of type-II analysis at a significant value of 0.05, respectively.References ADDIN EN.REFLIST Benjamini Y, and 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.Birney E, Clamp M, and Durbin R. 2004. GeneWise and genomewise. Genome Research 14:988-995.Castresana J. 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution 17:540-552.Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32:1792-1797.Loytynoja A, and Goldman N. 2008. Phylogeny-aware gap placement prevents errors in sequence alignment and evolutionary analysis. Science 320:1632-1635.Murrell B, Moola S, Mabona A, Weighill T, Sheward DJ, Pond SLK, and Scheffler K. 2013. FUBAR: A fast, unconstrained bayesian approximation for inferring selection. Molecular Biology and Evolution 30:1196-1205.Pond SLK, and Frost SDW. 2005. Datamonkey: rapid detection of selective pressure on individual sites of codon alignments. Bioinformatics 21:2531-2533.Pond SLK, and Muse SV. 2005. Site-to-site aariation of synonymous substitution rates. Molecular Biology and Evolution 22:2375-2385.Sunagar K, Johnson WE, O'Brien SJ, Vasconcelos V, and Antunes A. 2012. Evolution of CRISPs associated with toxicoferan-reptilian venom and mammalian reproduction. Molecular Biology and Evolution 29:1807-1822.Tamura K, Stecher G, Peterson D, Filipski A, and Kumar S. 2013. MEGA6: molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution 30:2725-2729.Wertheim JO, Murrell B, Smith MD, Pond SLK, and Scheffler K. 2015. RELAX: detecting relaxed selection in a phylogenetic framework. Molecular Biology and Evolution 32:820-832.Woolley S, Johnson J, Smith MJ, Crandall KA, and McClellan DA. 2003. TreeSAAP: selection on amino acid properties using phylogenetic trees. Bioinformatics 19:671-672.Yang ZH. 2007. PAML 4: Phylogenetic analysis by maximum likelihood. Molecular Biology and Evolution 24:1586-1591. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download