#!/usr/bin/perl
use strict;
use warnings;

my ($codes_dir,$name,$output_dir,$project)=@ARGV;
die "useage:perl $0 codes_dir name output_dir project" unless @ARGV==4;
my $snp="$output_dir/mutation/${name}.snp.Somatic.vcf.hg19_multianno.txt";
my $indel="$output_dir/mutation/${name}.indel.Somatic.vcf.hg19_multianno.txt";

open DRUG,"$codes_dir/info.txt";
<DRUG>;
my @genes;
while (<DRUG>){
        chomp;
        my @line=split(/\t/,$_);
        if ($line[0] eq $project){
                @genes=split(/\//,$line[2]);
        }
}

&pick_somatic($snp,"$output_dir/report/${name}.snp.Somatic.txt");
&pick_somatic($indel,"$output_dir/report/${name}.indel.Somatic.txt");

sub pick_somatic{
my ($in,$out)=@_;
open IN,"$in";
open OUT,">$out";
my $head=<IN>;
chomp $head;
print OUT join("\t",(split(/\t/,$head))[0..9]),"\tFreq","\tDP4","\tpredict_benign(MutationTaster/FATHMM/MetaSVM)\t",join("\t",(split(/\t/,$head))[10,11,16..20,23,28,32]),"\tSTR","\n";
while (<IN>){
        chomp;
        my @line=split(/\t/,$_);
	next unless(grep{$line[6] eq $_}@genes);
        next if ($line[11] !~/COS/  or $line[5]!~/exonic|splicing/ or $line[5]=~/ncRNA/ or $line[8] eq "synonymous SNV" or $line[17]>=0.01 or $line[18]>=0.01 or $line[19]>=0.01 or $line[20]>=0.01 or $line[23]>=0.01 or $line[28]>=0.01 or $line[32]>=0.01);
        my @var=split(/;/,$line[9]);
        my ($freq,$dp4)=(split(/:/,$line[-2]))[-2,-1];
        my $predict_benign=0;
        $predict_benign++ if ($line[50] eq "N" or $line[50] eq "P");
        $predict_benign++ if $line[56] eq "T";
        $predict_benign++ if $line[64] eq "T";
	next if $predict_benign==3;
        print OUT join("\t",@line[0..9],$freq,$dp4,$predict_benign,@line[10,11,16..20,23,28,32,$#line]),"\n";
        }
}