How to concatenate gene sequence?

Question

[buthercup_ch]

Hello,

With the aim of doing some phylogenetics based on Multi Locus Sequence Alignment, I need to generate concatenated sequences of several genes.

Does anyone how can I do it quickly by using any unix based command line? If yes, could you please help me and shae with me the program?

Thank you in advance

Example:

If I have 4 fasta files containing the sequence of 4 genes:

genea.fas

>geneA
ATGCCGTAATGCTAGCTAG

geneb.fas

>geneB
GTCGGCCTACGATCGGCCCTTTACG

genec.fas

>geneC
ACGCAGCCTGCGAGCTCA

gened.fas

>geneD
CATCCACTACCTTACGATCG

I need to obtain a final file containing 1 only concatenated sequence

concat.fas

>concat
ATGCCGTAATGCTAGCTAGATGCCGTAATGCTAGCTAGACGCAGCCTGCGAGCTCACATCCACTACCTTACGATCG

Answer 1

[nterhoeven]

This should work for you:

cat * | perl -ne 'BEGIN{print ">concat\n"}next if $_=~/^>/; chomp; print $_;' > concat.fa

This line concatenates all files in your directory and passes the output to perl. Then perl prints a header line (>concat) and every line (which does not start with >) without the newline.

Please try it on a test dataset and let me know, if it works for you ;-)

Answer 2

[iraun]

awk approach to combine sequences:

awk '/^>/ {if (c++ == 0) {print;}next}/^$/ {next}{printf "%s", $0}END {print ""}' input.fa > output.fa

The fasta header of the output sequence will be the the header of the first sequence in your input file.

Answer 3

[Anima Mundi]

A Python solution:

concat = ''

j = 0
while j < 4: # That is, the number of files you have.
    for line in open('gene' + chr(j + ord('a')) + '.fas'):
        if '>' not in line:
            concat += line
    j += 1

print '>concat\n' + concat.replace('\n', '')

Answer 4

[Matt Shirley]

cat *fas | grep -v '>' | tr '\n' '' > out.fas