Extract novel motifs from protein multiple sequence alignment

Question

[BioAnna]

Hello world!

I am new to R, and I would like some tips on motif detection in a protein multiple sequence alignment. I would like to extract the motifs using only some parts of the alignment (ex. aa 23-58, 123-567 etc).

I have already created a multiple sequence alignment of orthologues of this protein using MAFFT (browser version).

Some recommend to use MEME to create pssm.

So I would like some recommendations for packages, tools and any strategic tips for my research.

I will truly appreciate any help.

Thank you in advance.

[Gautier Richard]

So what you want to do is identifying motifs in protein sequences. But do you really need to align the proteins to get this information? Shouldn't MEME run on a fasta file where you have your different protein sequences? Couldn't a motif be at the end or the beginning of different proteins? Also, how many proteins do you want to analyse?

Also, consider that MEME is looking for rather long ungapped motifs, while DREME looks for short ungapped motifs. Perhaps for proteins it is better to use GLAM2 to find gapped motifs?

For more information: http://meme-suite.org/index.html

[BioAnna]

Thank you for your response. I have a group of 32 proteins, members of the same protein family. I'm trying to detect common de novo motifs (2-15 aa) in this protein group. So maybe GLAM2 is a better choice for this. I want to look in certain areas of the proteins, so I thought maybe if I could access GLAM2 from python or R, I could define better the search criteria.