Redundant Database for BLAST search

Question

[Emily]

Hi,

I have been running protien BLAST search on my protein which is present in many bacterial and eukaryotic species, as well as archaea. I am particularly interested in looking at which taxonomic groups have the protein versus those that have lost it.

The BLAST non-redundant database is grouping identical proteins with different taxonomic IDs so I am missing many accession codes from the hit table.

Does the non-redundant database group all identical protein as a single hit regardless of taxonomic group (or is it restricted to those with the same order or class etc?).

Is there a way of outputting the grouped proteins as separate entries in the hit and descriptions tables?

Is there a 'redundant' database that I search against locally?

Thanks,

Answer 1

[GenoMax]

Does the non-redundant database group all identical protein as a single hit regardless of taxonomic group

If the sequence is identical then yes the nr database fasta header can contain multiple entries (if sort of breaks the fasta format). See example below.

>WP_086194063.1 Re/Si-specific NAD(P)(+) transhydrogenase subunit alpha [Acinetobacter terrae] >OTG75608.1 NAD(P) transhydrogenase subunit alpha [Acinetobacter terrae] >WP_347200460.1 methylmalonyl-CoA mutase [Marivita sp.] >MEN8659902.1 methylmalonyl-CoA mutase [Marivita sp.] >MEN8682336.1 methylmalonyl-CoA mutase [Marivita sp.] >MEN8749575.1 methylmalonyl-CoA mutase [Marivita sp.]

You will have to post-process your output if you are interested in getting every entry in this "modified" fasta header.

Are you using -outfmt 6 in your output and is it only showing one ID? Otherwise that may be something to try.

Is there a 'redundant' database that I search against locally?

You can download the preformatted nr database files from https://ftp.ncbi.nih.gov/blast/db/ for local searches. Large database though so make sure you have enough resources.