delim2fasta.pl

#!/usr/bin/env perl
# Sam Shepard - 2016
# Delimited file to FASTA format. Attempts to deduct the sequence field.

use Getopt::Long;
GetOptions(
            'input-delimiter|I=s'  => \$inD,
            'output-delimiter|O=s' => \$outD,
            'strip-gap-chars|G'    => \$stripGaps,
            'jump-header-line|J'   => \$jumpHeader
);

if ( -t STDIN && !scalar(@ARGV) ) {
    $message = "Usage:\n\tperl $0 <delimited.txt> [options]\n";
    $message .= "\t\t-I|--input-delimiter <CHAR>\tParsing delimiter for input, default <tab>.\n";
    $message .= "\t\t-J|--jump-header-line\t\tSkip first line when processing.\n";
    $message .= "\t\t-O|--output-delimiter <CHAR>\tParsing delimiter for output header, default is pipe '|'.\n";
    $message .= "\t\t-G|--strip-gap-chars\t\tStrip gap characters found in sequences: '.-~:'\n";
    die( $message . "\n" );
}

if ( !defined($inD) ) {
    $inD = "\t";
}

if ( !defined($outD) ) {
    $outD = '|';
}

# discover sequence field
$/     = undef;
@lines = split( /\r\n|\n|\r/, <> );
chomp(@lines);
if ( scalar(@lines) == 0 ) { exit(0); }
if ($jumpHeader)           { $junk = shift(@lines); }
$firstLine = $lines[0];
@fields    = split( $inD, $firstLine );
$maxLen    = 0;
$seqField  = 0;

foreach $i ( 0 .. $#fields ) {
    $tmp = $fields[$i];
    $tmp =~ tr/ //d;
    if ( $tmp =~ /^[A-Za-z*.~:-]+$/ ) {
        if ( length($tmp) > $maxLen ) {
            $seqField = $i;
            $maxLen   = length($tmp);
        }
    }
}

# loop over the rest
foreach $line (@lines) {
    @fields = split( $inD, $line );
    $seq    = $fields[$seqField];
    $header = '';
    $seq =~ tr/ //d;
    if ($stripGaps) {
        $seq =~ tr/.:~-//d;
    }
    foreach $i ( 0 .. $#fields ) {
        if ( $i != $seqField ) {
            if ( $header ne '' ) {
                $header .= $outD . $fields[$i];
            } else {
                $header = $fields[$i];
            }
        }
    }

    if ( substr( $header, 0, 1 ) ne '>' ) { print '>'; }
    print $header, "\n", $seq, "\n";
}