How to calculate PSSM matrix for a protein?

Question

[Misha]

Hello,

How to generate PSSM for a protein sequence to form a feature vector using a window size technique in order to use it for machine learning purposes?

Thank you.

Answer 1

[Mensur Dlakic]

A simple way of doing that is to install BLAST locally, and run something like the following command:

psiblast -query protein.fas -db nr -num_iterations 3 -out_ascii_pssm protein.pssm

If you never ran BLAST locally, there is a detailed tutorial here.

In case you learn best by example, you may want to check out how the whole procedure is done by Porter5, in scripts directory.

[Misha]

Hi, Thanks a lot for the information. I will surely work on it.