how confident we are if a read aligned to 23S belong to the bacteria of interest and not other bacteria in the same region that are in our sample?

Question

[eli_bayat]

I have a somewhat maybe naive question here, my background is statistics and recently I entered Bioinformatics world and I enjoy every second of learning all about it. So my question is related to alignment:

I have a sample of different bacteria and I aligned it using bwa mem tool to bacteria references. I am interested in finding a mutation in a universal gene, say 23S or 16S in a specific bacteria, say gonorrhoeae. How confident we are on the result of the alignment? what if a read that aligned to 23S in gonorrhoeae genome comes from 23S of another bacteria and randomly placed in that region for gonorrhoeae. Is such scenario possible? is there a way to avoid or at least minimize this issue?

Thanks

[Mensur Dlakic]

It is not entirely clear to me what you are asking. For example, this part:

what if a read that aligned to 23S in gonorrhoeae genome comes from 23S of another bacteria and randomly placed in that region for gonorrhoeae.

Where did your sample come from? A pure culture or a metagenome?

Generally speaking, ribosomal RNA (rRNA) sequences are among the most conserved throughout the evolution, even between fairly unrelated organisms. This is to say that you may get 95% identity at rRNA level and still be talking about a different species, even a different genus. The closer you are to 100% identity, the more likely it is that you have at least a member of the same genus, though there is still no guarantee that you have the same species.

By the way, you could possibly use BLAST for this against rRNA reference database.