Identifying very closely related proteins

Question

[Baby Bioinformatician]

Hello Biostars, I have four dehydrogenases that exhibit stereospecificity for a compound. I am trying to identify the orthologs of these proteins in different genomes of streptomyces bacteria. However, since the four protein sequences are highly identical to each other, I get the same best hit for each of them when I perform a blast search in each genome. How best should I identify the orthologs of these proteins from my target genomes? Thank you for your suggestions.

Answer 1

[Mensur Dlakic]

If your proteins are highly identical to each other - what is the exact identity level? - their orthologs are highly identical as well. Therefore, it is not surprising that all of them have the same top hit. Sequence comparisons that detect orthology are not meant for detecting these fine differences between related proteins such as stereospecificity.

Though I doubt that this can be done properly by sequence comparisons, here is an idea: if you know the protein region that confers stereospecificity, rank your BLAST matches by identity/similarity in that region rather than total scores.