Create your own conference schedule! Click here for full instructions

Abstract Detail



Recent Topics Posters

Malley, Claire [1], Johnson, Matthew [2], Goffinet, Bernard [3], Shaw, Jonathan [4], Wickett, Norm [5].

A comparison of ortholog detection methods and their application to the moss phylogeny.

Bryophyta, mosses, are a diverse group of land plants, second only to angiosperms in number of species, that have undergone several, rapid radiations. Despite a rich history of systematic research, evolutionary relationships at some deep nodes and particularly among the pleurocarpous mosses remain unresolved. The identification of single-copy nuclear genes in mosses will be invaluable to address these relationships. Orthologous genes arise by speciation events, whereas paralogous genes arise by duplication, therefore it is more likely that orthologous gene trees will represent the species tree. With the support of the National Science Foundation's Assembling the Pleurocarp Tree of Life project, we present 26 new moss transcriptomes. We applied a custom bioinformatics pipeline with successive filters to remove sequences that are likely non-plant in origin; we then used these transcriptomes to compare the ortholog circumscription programs ProteinOrtho and OrthoFinder. Ultraconserved orthologs (UCOs) were identified using a subset of six moss transcriptomes generated for this project: Aulacomnium, Anomodon, Callicladium, Dicranum, Kindbergia, Leucobrum, and the proteome of the model moss Physcomitrella patens. Conserved, orthologous groups (orthogroups) were characterized by unchanging orthogroup membership across multiple programs and program parameters. Our first pass applied a conservative filter that requires the presence of a single transcript for each species. We consider this to be conservative due to the possibility that some genes may not be recovered in a transcriptome, and alternative splice forms (accounted for here, in part, prior to clustering) may result in the false rejection of one-to-one orthology. Hidden Markov Models (HMMs) were generated from a multiple sequence alignment of each orthogroup using the HMMER package, which we used to identify these UCOs in twenty additional transcriptomes. Orthogroups circumscribed with an independent clustering method (Yang & Smith 2014) were compared to those from ProteinOrtho and OrthoFinder. We used the UCOs and orthogroups to summarize how these methods differ in circumscribing multi-copy and low-copy gene families. Finally, we reconstructed a species tree using both supermatrix methods and methods that take gene-tree conflict into account. The orthogroup identities and HMMs generated here will likely be useful for future phylogenetic analysis in mosses.


Log in to add this item to your schedule

1 - Northwestern University, 633 Clark Street, Evanston, IL, 60208, United States
2 - Biology Department, 7033 1/2 N Sheridan Rd, Apt 2W, Chicago, IL, 60626, USA
3 - University Of Connecticut, Department Of Ecology & Evolutionary Biology, 75 N. Eagleville Road, U-3043, Storrs, CT, 06269-3043, USA
4 - 130 Science Drive, Box 90338, Durham, NC, 27708, USA
5 - Chicago Botanic Garden, 1000 Lake Cook Rd., Apt 2, Glencoe, IL, 60022, USA

Keywords:
Bryophyte
moss
Evolution
Phylogeny
Bioinformatics.

Presentation Type: Recent Topics Poster
Session: P
Location: Hall D/The Shaw Conference Centre
Date: Monday, July 27th, 2015
Time: 5:30 PM
Number: PRT047
Abstract ID:1833
Candidate for Awards:None


Copyright 2000-2015, Botanical Society of America. All rights reserved