Analysis of convergence between organismal traits and DNA/protein sequences
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.
R Corrected winsorization behavior Oct 10, 2018
data Added text to enrichment vignette Aug 23, 2018
ext Making vignette branch compatible with master Aug 9, 2018
inst/extdata Working towards a knittable version of the vignette Aug 9, 2018
man clickTool,xlimsForPlotRers Sep 27, 2018
src Adding makevars Apr 30, 2018
tests Fixed typo in tests Sep 18, 2018
vignettes Corrected winsorization behavior Oct 10, 2018
.DS_Store Knit enrichment vignette Sep 6, 2018
.Rbuildignore Latest updates to binary walk-through vitnette Nov 30, 2017
.gitignore~ Update the weight computation. Jan 3, 2018
DESCRIPTION ROxygenized version incorporating code updates Aug 10, 2018
LICENSE Create LICENSE Aug 31, 2017
NAMESPACE Update Sep 28, 2018


RERConverge is a set of software written in R that estimates the correlation between relative evolutionary rates of genes and the evolution of a convergent binary or continuous trait across a phylogeny.

Getting Started

Please refer to the wiki for detailed instructions to install RERConverge from scratch. For more information on running RERConverge, please see the full documentation (Link to R documentation) and R vignettes. (Link to R vignettes)

Quick Start


To run an analysis you will need:

  1. a trees file: a tab-delimited files with gene names and Newick format trees for each gene. Tree topologies must be the same for all genes, and at least one tree must contain all species in the dataset. We provide trees files for several clades here. (link to example datasets)

  2. information about phenotypes for species included in the dataset. For a binary trait analysis, this can either be in the form of: -a vector (in R) of species to include in the foreground.* -a tree object (made in R from a Newick tree) where branches are non-zero only for foreground lineages. *This may yield incorrect inferences for ancestral branches, so the tree object is preferred.

For a continuous trait analysis, this should be: -a named vector (in R) of quantitative phenotype values, where the names represent the species to which the phenotypes correspond.


Running RERConverge will produce the following outputs:

  1. an object containing, for each gene, the correlation between its relative evolutionary rate and the trait of interest, along with the estimateed p-value
  2. an object containing, for each gene, its relative evolutionary rate for each branch of the phylogeny, which can be used in the included visualization scripts (e.g., to illustrate the difference in relative evolutionary rate between foreground and background branches)


See also the list of contributors who participated in this project.


RERConverge can be cited as follows:

For coding sequences:
Chikina M, Robinson JD, Clark NL. Hundreds of Genes Experienced Convergent Shifts in Selective Pressure in Marine 
Mammals. Mol Biol Evol. 2016;33: 2182–92. doi:10.1093/molbev/msw112

For conserved non-coding sequences:
Partha R, Chauhan B, Ferreira Z, Robinson J, Lathrop K, Nischal K, et al. Subterranean mammals show convergent 
regression in ocular genes and enhancers, along with adaptation to tunneling. In press eLife. 


This project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details


  • Methods for computing weights rely on ideas from the following paper:
Law CW, Chen Y, Shi W, Smyth GK. voom: Precision weights unlock linear model analysis tools for RNA-seq read
counts. Genome Biol. 2014;15: R29. doi:10.1186/gb-2014-15-2-r29

  • Projection operations are drawn from the following paper:
Sato T, Yamanishi Y, Kanehisa M, Toh H. The inference of protein-protein interactions by co-evolutionary 
analysis is improved by excluding the information about the phylogenetic relationships. Bioinformatics. 
Bioinformatics Center, Institute for Chemical Research, Kyoto University, Gokasho, Uji, Kyoto 611-0011, 
Japan.; 2005;21: 3482–3489. doi:10.1093/bioinformatics/bti564