Entering edit mode
7.8 years ago
SHAD
▴
10
How to select homologous sequences from BLAST search to generate multiple sequence alignment? I mean how many sequences, what is the best E-value cut-off for these sequences.
There isn't a fixed value for this. Take a look at: http://biology.stackexchange.com/questions/19338/e-value-blast-cut-off
One could generate an MSA for a short stretch of AA's that includes a domain (or some other converved sequence) or you may be referring to a full length protein.
Thank you for you response, OK so If I want to identify conserved regions I should go with smaller E-value right?
In general, yes. But keep the caveats defined in the stackexchange post in mind as you go about doing this.