#!/usr/bin/perl -w
# NCBI_condense_names.pl  2006 NUF
# Replaces entry names with a more simple name
#
# Enter at command line: condense_NCBI_names.pl <input file> <output file>
#
# Example:
#   >gi|68551093|ref|ZP_00590521.1| Rhodanese-like [Pelodictyon phaeoclathratiforme BU-1]
# becomes:
#   >ZP_00590521_Pelodictyon_phaeoclathratiforme_BU1
#
# This script also replaces some genus names with abbreviations (see end of script)
#
# Modified by S.R. Santos 05 Dec 2007
#
# NOTE: this script is good, but cannot take into account every possible iteration in a descriptor line, so your mileage may vary 


$inputfile=$ARGV[0];
$outputfile=$ARGV[1];

open(FH,$inputfile) or die "Cannot open $inputfile.\n";
@line_list=<FH>;
chomp @line_list;
close(FH);

open(FH,">$outputfile") or die "Cannot open $outputfile.\n";

foreach $line (@line_list) {
  if ($line =~ />/ ) {
#   Replace everything up to the first "[" with the accession number
    for (1..3) {
      $a=index($line,"|"); substr($line,1,$a)=""; # Remove first 3 fields
    }
    $a=index($line,"|");
    $a2=index($line,".");
    if (($a2>-1) && ($a2<$a)) { $a=$a2; }
    $b=index($line,"[");
    substr($line,$a,$b-$a+1)="_";
#   Remove everything starting with ":" or "]" or "/" to the end
    foreach $a (qw/ : ] \/ /) {
      if ($line =~ /$a/ ) {
        $line=substr($line,0,index($line,$a));
      }
    }
#   Remove characters and words (max 10 occurrences of each can be removed, can be modified for specific cases)
    for (1..10) {
      foreach $a (qw/ sp. strain isolate chromosome \( \) \. /) { $line =~ s/$a//; }
    }
#   Replace double spaces with single space
    $ss=" "; $ds="  ";
    for (1..10) { $line =~ s/$ds/$ss/; }
#   Replace all " "  with "_" and remove "-"
    for (1..10) { $line =~ s/$ss/_/; $line =~ s/-//; }
#   Remove any terminal "_"
    substr($line,-1) =~ s/_//;
    
#   Condense genus names
    $line =~ s/Chlorobium/Chl/;
    $line =~ s/Pelodictyon/Pld/;
    $line =~ s/Prosthecochloris/Ptc/;

  }
  print FH "$line\n";
}

close(FH);