RNA-Seq data analysis mapping to closest genome as reference genome

Question

[ramzz.1987]

Can I use the closest genome as the reference genome for RNA-Seq analysis instead of Denove genome assembly. For example, I have a genome with strain A and can I use the same genome with strain B as my reference genome. Please give your feedback if it makes sense biologically.

[Medhat]

Is it possible that you can assemble the transcript?

Answer 1

[unawaz]

It depends on how closely related the two organisms are. Otherwise you can perform a de novo transcriptome assembly using tools like Trinity (https://github.com/trinityrnaseq/trinityrnaseq/wiki), and then BLAST it against the genome of the closely related one.

[ramzz.1987]

Actually, I want to perform Reference based analysis and when I tried mapping to closest genome I found around 80% alignment rate. But I want to make sure does this analysis make biological sense.

[Charles Warden]

You might see an 85-90% alignment rate for some protocols aligned against a well-characterized genome (such as human or mouse).

So, in that context, an 80% alignment rate wouldn't seem too bad, and having the more thoroughly reviewed annotations might be helpful (versus trying to work with the Trinity assembly). However, for your particular library and analysis strategy, I don't know what the alignment rate against a complete, published genome would be.

[ramzz.1987]

Hi Dr. Charles,

Thank you very much for your reply. It is a bacterial published complete genome. We have reviewed annotations and seems to be good.

[Charles Warden]

If you use TopHat, perhaps you check the unaligned reads for contamination? For example, you could try doing an assembly of the unaligned reads and then BLAST some large and/or high coverage contigs.

If you had a pure sample for a bacterial genome with relatively low population variation (and/or minimal regions that are hard to assembly for the genome sequence), an 80% RNA-Seq alignment may be low. However, it is hard for me to say for certain.