How to filter seqs by their 1st nt

[cascoamarillo]

Hi,

It sounds silly, but I couldn't find an approach for filter/separate sequences (in a fasta file) according to their first (or specific motif) nucleotide; I mean, to generate a file of seqs that starts all of them with the same nucleotide (A, T, G or C). Any ideas?

regards

[NicoBxl]

simple with cat and grep :

here's the code to filter only the sequences that start with an A. For the other nt, juste change the A into T,G or C. The only thing, it's that each sequence must be written in one line. Fasta Width from galaxy do that (http://main.g2.bx.psu.edu/) Fasta Manipulation > Fasta Width.

cat input.fasta | grep ^A -B 1 | grep '\-\-' -v > output.fasta

[cascoamarillo]

Thanks mate!

It works fine.