I have population genomic sequence data of several individuals and I suspect that the coalescence times vary across the genes and genomic regions (due to introgression and recombination). Estimating the coalescence times using a Bayesian approach is too slow for these data because of the many regions I want to know the divergence time of, even on a cluster. Is there a good alternative that I can use?
To estimate the divergence of the sequences, Vector NTI Advance is a license software from Life Technologies with CAP 3 Algorithm. The software has 5 modules to plan, design, perform, and organize your sequence analysis and Genomic work. Mega5 is a free software available online. ClustalX/W, PRRP, DIALIGN, HMMT and BLOSUM62 is another algorithms to try. Good Luck.
Well... there is no such thing as a molecular clock and certainly not for protein coding sequences.
Trying to estimate divergence time is probably not a good idea. Even if a phylogenetic tree can provide successive ancestors, the length of an internal branch cannot simply be used to estimate a duration (see for example molecular trees and branch lengths, and correlation with a fossil record when available).
You can use Phylogenetic Network Software. http://www.fluxus-engineering.com/sharenet.htm.
But the best way is using the Bayesian approach "BEAST".
The above mentioned softwares are good but not accurate. If you can merge or compare both the bioinformatics methods with fossil records, you will have very interesting article to publish. This just an idea.
Also PhyloBayes do calibrations. And of course CodeML (maximum likelihood inference).
- Badges
- Science method