HyPhy message board | |
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl
Methodology Questions >> How to >> Co-evolution Detection Using Distance Matrices http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1229815270 Message started by Andrew_Roth on Dec 20th, 2008 at 3:21pm |
Title: Co-evolution Detection Using Distance Matrices Post by Andrew_Roth on Dec 20th, 2008 at 3:21pm
I am interested in trying to detect co-evolution between two sets of genes that are thought to interact. I have been reading the literature on this and it seems that the most popular method is to build distance matrices for each set of genes. The distance matrices can then be checked for correlations using a number of different metrics.
I was wondering if anyone has worked with this sort of problem and could shed some light on what the best distance measure would be to construct the matrices. The most popular choice seems to be to use a simple distance based on sequence identity. The major problem with this method is that it can't distinguish correlations due to phylogeny versus those due to co-evolution. I am contemplating using the dN/dS values as a distance measure since this might be a more reliable way to detect co-evolution. I would expect that strongly correlated dN/dS values would be indicative of genes evolving under similar selective regimes which is suggestive of co-evolution. Does anyone agree with that, or is there be some more appropriate measure I should use? Thank you for any help. Andrew |
Title: Re: Co-evolution Detection Using Distance Matrices Post by kevintz on Jan 13th, 2009 at 7:21am
host and pathogen coevolution is really interesting. I am working on microbiology, and always intent to find a way to touch on the field.
Thanks. |
HyPhy message board » Powered by YaBB 2.5.2! YaBB Forum Software © 2000-2024. All Rights Reserved. |