gistfile1.pl

#!/usr/bin/env perl
use 5.010;
use strict;
use List::MoreUtils qw(pairwise);
use Text::Extract::Word;

sub foo {
    sprintf "%.3d", shift;
}

sub handle_section {
    my $str = shift;
    my ( $gene_name, $_ ) = $str =~ /(.*?\n)((.*\n)*)/;
    s/\n//g;

    ($gene_name) = $gene_name =~ />\s(.*?)\s/;

    my ( $icont, @iarray, @earray, @temp );
    while (/[a-z]+|[A-Z]+/gc) {
        if ( substr( $&, 0, 1 ) ~~ @{ [ 'a' .. 'z' ] } ) {
            push @iarray, 'I' . foo( ++$icont ) . $&;
        }
        else {
            push @temp, $&;
        }
    }

    @earray = map {
            "E"
          . foo( ( $_ + 1 ) . ( $_ + 2 ) )
          . substr( $temp[$_], ( length $temp[$_] ) - 10 )
          . substr( $temp[ $_ + 1 ], 0, 10 )
    } ( 0 .. $#temp - 1 );

    say $gene_name;
    say for pairwise { ( $a, $b ) } @iarray, @earray;
    print "\n";
}

for(glob "*.doc"){
    return unless /\.doc/;

    my $file = Text::Extract::Word->new($_);
    my $text = $file->get_text();

    open my $in, "<", \$text;
    open STDOUT, ">", "$_.out.txt";

    my $section = "";

    while (<$in>) {
        if (/^\s*$/) {
            handle_section($section);
            $section = "";
        }
        else {
            $section .= $_;
        }
    }
}