pairwise alignment with multiple sequences in a fasta file to a reference sequence.

Question

[kanshenglong]

Hi all, I hope that a script or software to do alignment the sequence in a multiple sequences file to a reference sequence, not do multiple alignment of nucleic acid and protein sequences. Best!

[GenoMax]

You can use one of these pair-wise alignment programs from EMBOSS. Download EMBOSS software here.

As for a shell script you could use something like this (pseudocode)

for f in *.fasta
do
    your_choice_of_program reference.fa $f output_options_for_your_program
done

[James Blachly]

Agree with genomax2's answer, with the additional qualification that it depends whether you want to do local alignment or global alignment.

EMBL-EBI pairwise alignment tools

If you think some of your isoforms may be 5' or 3' incomplete, you may get better results with local alignment, depending on how much sequence is missing at the terminii.

[microfuge]

In case you mean a mapping like alignment. BLAT seems like a good choice. Its available as a standalone software as well.

Answer 1

[piet]

You can use 'bwa mem' to map a bunch of sequences to a reference sequence. 'bwa mem' can read FASTA files directly, you do not have to convert them to FASTQ.

bwa index chromosome.fasta
bwa mem chromosome.fasta lots_of_sequences.fasta

This can be handsome to map a collection of genes (coding sequences) to a newly assembled chromosome rather quickly.

A: Contig alignment software

Answer 2

[James Blachly]

(Changing my comment to an answer)

Agree with genomax2's answer, with the additional qualification that it depends whether you want to do local alignment or global alignment.

EMBL-EBI pairwise alignment tools

If you think some of your isoforms may be 5' or 3' incomplete, you may get better results with local alignment, depending on how much sequence is missing at the terminii.

The bottom line is that you can get very different results with global vs local alignment, so one will likely perform better than the other depending on the nature of your sequences; you may need to try both if you are not sure.