Pairwise comparison of AA fasta files which contain multiple domains

Question

[MaheJaan]

I have about 5000 .fasta files which contain 2 versions (transcript) of the same gene split into their domains, like so:

 >zf-C4_1_ENST00000512784
RLCLVCGDIASGYHYGVASCEACKAFFKRTIQGNIEYSCPATNECEITKRRRKSCQACRF
MKCLKVGMLK

>Hormone_recep_1_ENST00000512784
IKALTTLCDLADRELVVIIGWAKHIPGFSSLSLGDQMSLLQSAWMEILILGIVYRSLPYD
DKLVYAEDYIMDEEHSRLAGLLELYRAILQLVRRYKKLKVEKEEFVTLKALALANSDSMY
IEDLEAVQKLQDLLHEALQDYELSQRHEEPWRTGKLLLTLPLLRQTAAKAVQHFYSVKLQ

>zf-C4_1_ENST00000644823
RLCLVCGDIASGYHYGVASCEACKAFFKRTIQGNIEYSCPATNECEITKRRRKSCQACRF
MKCLKVGMLK

>Hormone_recep_1_ENST00000644823
IKALTTLCDLADRELVVIIGWAKHIPGFSSLSLGDQMSLLQSAWMEILILGIVYRSLPYD
DKLVYAEDYIMDEEHSRLAGLLELYRAILQLVRRYKKLKVEKEEFVTLKALALANSDSMY
IEDLEAVQKLQDLLHEALQDYELSQRHEEPWRTGKLLLTLPLLRQTAAKAVQHFYSVKLQ

I will like to run a pairwise comparison on the same domains.

When I try to run this with msa(mySequences) it is treated as 4 sequences to compare, not just the domains. Any help on how I can so this, and maybe make a loop in R for the other 5000 files?

Answer 1

[Mensur Dlakic]

You can use this script to sort your sequences by size, since there seem to be a clear difference in domain sizes. After that split the resulting file into the two domains, and proceed as planned.

The syntax:

sort_contigs.pl -b -z -p your_file.fas sorted_file.fas