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

my ($output_dir,$name)=@ARGV;
die "useage:perl $0 output_dir tumor" unless @ARGV==2;

open IN,"$output_dir/mutation/${name}.snp.indel.Somatic.annoall.hg19_multianno.txt";
my $head=<IN>;
open OUT,">$output_dir/mutation/${name}.snp.indel.Somatic.annoall.hg19_multianno_filtered.txt";
print OUT "可信\t$head";

while(<IN>){
	chomp;
	my @line=split(/\t/,$_);
        my $freq=(split(":",$line[-2]))[-5];
	my $FREQ=$freq=~s/%//r;
	next if $line[9] eq '.';
	my @hgvs=split(/,/,$line[9]);
	my $hgvs=$hgvs[0];
	$hgvs=~/(\S+):(\S+):exon(\d+):c\.(\S+):p\.(\S+)$/;
	my $gene=$1;
	if($line[8] ne "synonymous SNV" and $line[8] ne "unknown" and $line[17]<0.01 and ($line[$#line]!~/STR/ or (length($line[3])>=4 or length($line[4])>=4) or
        $FREQ>5) and $line[18]<0.01 and $line[19]<0.01 and  $line[20]<0.01 and $line[23]<0.01 and $line[28]<0.01 and $line[32]<0.01){
		if (my $transcript=&transcript($gene)){					
			if(grep{/$transcript/}@hgvs){
				$hgvs=(grep{/$transcript/}@hgvs)[0];
			}
		}
		$line[9]=$hgvs;
		print OUT "1\t",join("\t",(@line[0..4],"exonic",$gene,@line[7..$#line])),"\n";
	}
}

##$line[100] eq 'PASS'

open IN2,"$output_dir/mutation/${name}.snp.indel.Germline.anno.hg19_multianno.txt";
my $head=<IN2>;
open OUT2,">$output_dir/mutation/${name}.snp.indel.Germline.anno.hg19_multianno_filtered.txt";
print OUT2 "临床意义\t$head";

while(<IN2>){
        	chomp;
		my @line=split(/\t/,$_);
		my $freq=(split(":",$line[-2]))[-2];
		my $FREQ=$freq=~s/%//r;
		next if $line[9] eq '.';
        	my @hgvs=split(/,/,$line[9]);
        	my $hgvs=$hgvs[0];
        	$hgvs=~/(\S+):(\S+):exon(\d+):c\.(\S+):p\.(\S+)$/;
        	my $gene=$1;
		next if ($FREQ<10);
        	if($line[8] ne "synonymous SNV" and $line[8] ne "unknown" and $line[18]<0.01 and $line[19]<0.01 and  $line[20]<0.01 
        and $line[23]<0.01 and $line[28]<0.01 and $line[32]<0.01 and $line[16]=~/pathogenic|Affects|association|Conflicting|sensitivity|drug|other|risk|protective|Uncertain|not_provided|\./i){
		if (my $transcript=&transcript($gene)){					
			if(grep{/$transcript/}@hgvs){
				$hgvs=(grep{/$transcript/}@hgvs)[0];
			}
		}
		$line[9]=$hgvs;	
		if($line[16]=~/Likely_pathogenic|drug/i){
			print OUT2 "2\t",join("\t",(@line[0..4],"exonic",$gene,@line[7..$#line])),"\n";
		}elsif($line[16]=~/pathogenic/i and $line[16]!~/Conflicting/i){
			print OUT2 "1\t",join("\t",(@line[0..4],"exonic",$gene,@line[7..$#line])),"\n";
		}else{
			print OUT2 "3\t",join("\t",(@line[0..4],"exonic",$gene,@line[7..$#line])),"\n";
		}
	}
}


sub transcript{
	my $gene=shift @_;
	open TR,"/dataseq/jmdna/codes/reportbase/oncokbgene.txt";
	my %oncogene;
	while(<TR>){
		chomp;
		my @line=split;
		$oncogene{$line[0]}=$line[2];			
	}
	if (exists $oncogene{$gene}){
		$oncogene{$gene}=~s/\.\d+//;
		return $oncogene{$gene};
	}else{
		print "$gene has no NM id in oncokbgene.txt";
		return "";
	}
}