convert detailed alignment format to fasta

Question

[sapuizait]

Dear all

Apologies if this has been asked before but I cannot find a useful answer around

I am using kma to align reads against a set of genes and I am looking for a method/script/software to convert the detailed alignment format to a fasta format:

example input format:

# VFG000863(gb|BAA94855) (astA)
template:   ATGCCATCAACACAGTATATCCGAAGGCCCGCATCCAGTTATGCATCGTGCATATGGTGC
            ||||||||||||||||||||_||_||_|||_|||||||||||||||||||||||||||||
query:      ATGCCATCAACACAGTATATTCGGAGACCCACATCCAGTTATGCATCGTGCATATGGTGC

template:   GCAACAGCCTGCGCTTCGTGTCATGGAAGGACTACAAAGCCGTCACTCGCGACCTGA
            |||||||_|||||||||||||||||||||||||||||||||||||||||||||||||
query:      GCAACAGTCTGCGCTTCGTGTCATGGAAGGACTACAAAGCCGTCACTCGCGACCTGA

# VFG000924(gb|NP_752610) (fepB)
template:   GTGAGACTCGCCCCGCTCTACCGCAACGCCCTTCTATTAACAGGACTTTTGCTTTCAGGA
            ||||||||||||||||||||||||||||||||||||||||||||||||||_|||||||||
query:      GTGAGACTCGCCCCGCTCTACCGCAACGCCCTTCTATTAACAGGACTTTTACTTTCAGGA

template:   ATAGCCGCAGTTCAGGCCGCCGACTGGCCGCGTCAGATTACTGACAGCCGTGGCACTCAT
            ||||||||||||||||||||_|||||||||||||||||||||||||||||||||||_|||
query:      ATAGCCGCAGTTCAGGCCGCTGACTGGCCGCGTCAGATTACTGACAGCCGTGGCACACAT

thanks

[Pierre Lindenbaum]

an example of output is needed.

[sapuizait]

sorry, output would look like this:

>VFG000863(gb|BAA94855) (astA) template
ATGCCATCAACACAGTATATCCGAAGGCCCGCATCCAGTTATGCATCGTGCATATGGTGCGCAACAGCCTGCGCTTCGTGTCATGGAAGGACTACAAAGCCGTCACTCGCGACCTGA

>VFG000863(gb|BAA94855) (astA) query
ATGCCATCAACACAGTATATTCGGAGACCCACATCCAGTTATGCATCGTGCATATGGTGCGCAACAGTCTGCGCTTCGTGTCATGGAAGGACTACAAAGCCGTCACTCGCGACCTGA

>VFG000924(gb|NP_752610) (fepB) template
GTGAGACTCGCCCCGCTCTACCGCAACGCCCTTCTATTAACAGGACTTTTGCTTTCAGGAATAGCCGCAGTTCAGGCCGCCGACTGGCCGCGTCAGATTACTGACAGCCGTGGCACTCAT

>VFG000924(gb|NP_752610) (fepB) query
GTGAGACTCGCCCCGCTCTACCGCAACGCCCTTCTATTAACAGGACTTTTACTTTCAGGAATAGCCGCAGTTCAGGCCGCTGACTGGCCGCGTCAGATTACTGACAGCCGTGGCACACAT

[cpad0112]

with seqkit and sed:

$ cat <(sed -r '/template|#/!d; s/#/>/; s/^\w+\W\s+//;/>/ s/$/ template/' test.fa) <(sed -r '/query|#/!d; s/#/>/; s/^\w+\W\s+//;/>/ s/$/ query/' test.fa) | seqkit -w 0 sort --quiet -n

>VFG000863(gb|BAA94855) (astA) query
ATGCCATCAACACAGTATATTCGGAGACCCACATCCAGTTATGCATCGTGCATATGGTGCGCAACAGTCTGCGCTTCGTGTCATGGAAGGACTACAAAGCCGTCACTCGCGACCTGA
>VFG000863(gb|BAA94855) (astA) template
ATGCCATCAACACAGTATATCCGAAGGCCCGCATCCAGTTATGCATCGTGCATATGGTGCGCAACAGCCTGCGCTTCGTGTCATGGAAGGACTACAAAGCCGTCACTCGCGACCTGA
>VFG000924(gb|NP_752610) (fepB) query
GTGAGACTCGCCCCGCTCTACCGCAACGCCCTTCTATTAACAGGACTTTTACTTTCAGGAATAGCCGCAGTTCAGGCCGCTGACTGGCCGCGTCAGATTACTGACAGCCGTGGCACACAT
>VFG000924(gb|NP_752610) (fepB) template
GTGAGACTCGCCCCGCTCTACCGCAACGCCCTTCTATTAACAGGACTTTTGCTTTCAGGAATAGCCGCAGTTCAGGCCGCCGACTGGCCGCGTCAGATTACTGACAGCCGTGGCACTCAT

seqkit is for sorting the sequences by header and for printing sequence in a single line.

[sapuizait]

Thanks! Very cool use of sed!