This is a test version of Biostars. For the public version, visit https://www.biostars.org.
What Needs Needs To Be Parallelized In Phylogenetics?

Hello.

I'm in a masters program focusing on data mining and high performance computing (I have a computer science background). I'm interested in bioinformatics and particularly in phylogenetics and I was wondering if there are any algorithms or applications that bioinformaticians would benefit from being parallelized?

For example, I understand MrBayes splits MCMC chains using MPI, but also know that it only does Bayesian inference and not Maximum Likelihood. Are there holes in parallelized toolsets which I could potentially try to fill?

Sorry if this post sounds ignorant. I'm only beginning to learn about bioinformatics and I'm exploring potential thesis topics. All comments are welcome!

phylogenetics data

3 answers

Co-phylogenetic analyses haven't been paralllelized and very little work has been done in the field recently.

RAxML does parallel computing for maximum likelihood inference of (large) phylogenetic trees. In their own words they implemented "fine-grained parallelism that can be exploited on shared memory machines or multi-core architectures and coarse-grained paral- lelism that can be exploited on Linux clusters." They have MPI and Pthreads versions of the program.

You can read more on that and get all the references you need in their manual, available here.

Phylogenetic analysis usually start with a multialignment and I think you will find that most (if not all!) multialignment programs are not parallelized. I am sure work on that would be highly appreciated.

Thanks for the FastPhylo mentioning Roman! But FastProt, as you may know, has an MPI implementation and it will be included in the next release (which is being prepared).

Good to know! Thanks.

Log in to answer this question.