Need project Ideas in Molecular Phylogenetics

Question

[theblackbiologist]

So I have a project student (undergraduate) who has been assigned a project to study the evolution of a gene (and possibly related genes) through species. I am not an evolutionary biologist, and hence am not too experienced in designing a project, and hence I thought I'd run by some ideas by the fine folk here. The only limitation of this project is that it should be dry-lab (bioinformatics based)

The first aim is obviously Sequence data collection for the requisite gene (and a few for a good outgroup), plot them n a phylogenetic tree, and analyse the distances, Ka and Ks ratio.

The next thing I thought he could do was a simple data collection for SNP analysis, and see if the SNP's are conserved, across species, and possibly if they impact protein structure.

Any other ideas? Don't be afraid to say something that is obvious to you, as that may not be the case for me, as I have written down all the ideas in my head over here.

Answer 1

[madkitty]

Here is a list of ideas, it could be interesting especially if the gene of interest is long:

• Looking for non-synonymous SNPs if any and see how it affects the protein structure (you can predict up the tertiary structure).
• Looking for short and long indels and see if they are frameshifting
• Looking at the whole genome? Why just one gene? And draw phylogenetic tree (bayesian inferences with an outgroup + likelihood on each branch)
• Multiple alignment across different related species to summarize differences
• Depending on the hypothesis and the output of the sequencing you can question many things, is there any translation from another genome/specie (eukaryotic/prokaryotic), EST, etc.. (Though you're limiting the possibilities when it's just one gene)

Contact me in PM if you have any specific questions I can help you with. Good luck :)

[theblackbiologist]

The gene is quite short about 14kb, the transcript is about 3 kb, and the protein product is about 372aa. The reason for a single/related genes is the simple reason that we are a lab that is planning to use that gene of interest (MSTN) somewhere down stream, and hence it's be good to know how it came to be (plus it's a really beautiful story on how nature selected for negative muscle size). Thanks for the ideas, I'll chew on them. A few quick questions though

I can ask him to look for the SNPs (and by extension other INDELS) and see how it affects the protein structure (I presume Phyre ?) Do you have a good workflow for this ?

[madkitty]

You might want to chat by email regarding the workflow, you have different options that depend on his level of knowledge in bioinformatics and how much you want to teach him in this regard. You can contact me at [e-mail] (I'm a research assistant in genome sequencing)

[theblackbiologist]

Sent you an email. What methods you suggest that are not too complex, but gives good results? I am willing to hold the students hand for some time.