Identify homolog sequences that appear in multiple RefSeq files

Question

[sam]

Hello,

I have a set of 4 RefSeq fasta files belonging to 4 different species. I'm trying to identify homologs that appear in all 4 RefSeq files. Whats the best and most efficient way to do that? Please note that I only want to identify sequences that are homologous in all 4 files.

[Sishuo Wang]

OrthoMCL might be helpful.

It might take a long time to run the whole analysis for all of your species.

Answer 1

[Siva]

Depending on the species in your dataset, you can use any of the precomputed resources from NCBI

• Protein Clusters
• Homologene
• COG

These resources provide FTP files that have either protein GI or accessions.

Or you could use one of the sequence clustering tools (BLASTClust, UCLUST or CD-HIT), though they can be less accurate.