How to blastp

Question

[frenchmytoast112]

Hello everyone,

I am looking to perform blastp with a few sequences on a few hundred genomes using the nr database.

My progress thus far involves:

1. I know there is the NCBI website for blastp. However, I am not sure how I can make the input for many genomes. I know the taxid, gi and acc for my genomes but I would not rather type the taxid for each of my genomes into the input bar (for the nr database).
2. I found out I can use the blast+ suite on my Ubuntu. Again, I am not sure how to use it. I found some tutorials to create my own databases, but since I have low computational power on my laptop I would rather like to send my files to the NCBI server. So how can I do properly?

Thanks a lot in advance!

[Jean-Karim Heriche]

You could write a script to send your queries to the NCBI server. BioPerl has Bio::Tools::Run::RemoteBlast for this or you can roll your own using the NCBI Blast API. Just be careful not to hammer their server or you may get banned from it.

Answer 1

[dago]

If you want to perform a blastp search for the protein encoded in you genome, you can use the .faa of your genomes as -query in blasp and select nr as -db. To perform search on the NCBI server you can select the flag -remote.

See blast option here.

[frenchmytoast112]

What is the difference between the genomes in my .faa and the nr database? Shouldnt first download the genome from the nr database? And my sequence is from a separate genome and has nothing to do with the ones I am trying to blast.