This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Analyzing A Genome?

Dear Biostar people,

I just started a class in college. It is a small class <10 people. In it, we are given the entire text sequence of a new prokaryotic organism whose genome has never before been seen. We are asked to analyze it's genome, but given almost no resources or guidance. Could anyone tell me the best way to get started, scripts to run, techniques to apply, and the best ways to document my findings?

Thank you so much.

sequence-analysis homework

Your question is too general. First of all, I will disenroll myself from a class where instructor asks you to analyze a genome but doesn't give you any resource or guidance, Go to Pubmed and read some papers. This is how I learnt about Next Generation sequencing. There is no cookbook to NGS analysis. It is all about reading papers and hands on. I have been doing it for 3 years now but still don't know everything of it.

as OP says, the class just started, it is too early to make that decision - to drop a class based on this question, assuming that OP put time and effort into semester schedule planning and so on. it might be that instructor wasn't clear, or was misinterpreted by OP.

If it was the first class, it might well be a didactic move to see what students come up with by themselves, providing a little room for creativity and own initiative. Why does everything have to come pre-packaged and digested in the first place? Ofc, if the course stays like this, it is possibly not too good (what would be the purpose of a course without any guidance or advice?), but please wait and see. Certainly, to find biostar was a smart move ;)

I'm guessing the teacher is testing the class or giving them a hard time for fun (I'll do that)

Welcome to Biostars!
First of all, I would ask my professor about this. What does he/she mean by "analyzing"? After that it will be easier to help you.

3 answers

The first things I would do with an unknown prokaryotic genome :

  • find Open Reading Frames
  • find similarities: BLAST (a sliding window of) your genome to other known genomes.

Hope this doesn't sound stupid, but what about finding and comparing rRNA genes to known species?

For non-coding genes (such as rRNA) you can use aligments to a close species or prediction with RFAM/Infernal http://infernal.janelia.org/

For starters, find out what other bacteria it is related to. BLASTing some random sequences against nr is not a bad place to start there. Or you could get rRNA sequence (like the 16S sequence) from some random bacterium, BLAST that against your genome to identify your bug's rRNA, then BLAST that against nr.

Then, as Pierre suggested, you want to find an online program that will find ORFs, and you want to find some online program that will translate those ORFs, and tell you what genes they resemble. That will tell you something about what your bug can do.

You could Blast2Go and get some inital annotation and then do a phylogenetic tree using the whole genome or selected genes.

ORF finding is a must.

You could use your own scripts, ready-made tools or just a browser such as Artemis to get GC content and numerous other statistics about that genome.

Some of those statistics and graphs in Artemis can help you easily identify genes, GC islands and other features.

After that you could use a tool such as ACT and a pairwise blast with some of the most similar (and well annotated) organisms to transfer annotation info or blast-em and see what you get.

Cheers,

IV

Log in to answer this question.