&usage if (@ARGV < 1);
my $command = shift(@ARGV);
my %func = (subsam=>\&subsam, listsam=>\&listsam, fillac=>\&fillac, qstats=>\&qstats, varFilter=>\&varFilter,
- hapmap2vcf=>\&hapmap2vcf);
+ hapmap2vcf=>\&hapmap2vcf, ucscsnp2vcf=>\&ucscsnp2vcf);
die("Unknown command \"$command\".\n") if (!defined($func{$command}));
&{$func{$command}};
}
}
sub qstats {
- my %opts = (r=>'', s=>0.01);
- getopts('r:s:', \%opts);
+ my %opts = (r=>'', s=>0.02, v=>undef);
+ getopts('r:s:v', \%opts);
die("Usage: vcfutils.pl qstats [-r ref.vcf] <in.vcf>\n
Note: This command discards indels. Output: QUAL #non-indel #SNPs #transitions #joint ts/tv #joint/#ref #joint/#non-indel \n") if (@ARGV == 0 && -t STDIN);
my %ts = (AG=>1, GA=>1, CT=>1, TC=>1);
my %h = ();
+ my $is_vcf = defined($opts{v})? 1 : 0;
if ($opts{r}) { # read the reference positions
my $fh;
open($fh, $opts{r}) || die;
while (<$fh>) {
next if (/^#/);
- $h{$1,$2} = 1 if (/^(\S+)\s+(\d+)/);
+ if ($is_vcf) {
+ my @t = split;
+ $h{$t[0],$t[1]} = $t[4];
+ } else {
+ $h{$1,$2} = 1 if (/^(\S+)\s+(\d+)/);
+ }
}
close($fh);
}
my @s = split(',', $t[4]);
$t[5] = 3 if ($t[5] < 0);
next if (length($s[0]) != 1);
- push(@a, [$t[5], ($t[4] eq '.' || $t[4] eq $t[3])? 0 : 1, $ts{$t[3].$s[0]}? 1 : 0, $h{$t[0],$t[1]}? 1 : 0]);
+ my $hit;
+ if ($is_vcf) {
+ $hit = 0;
+ my $aa = $h{$t[0],$t[1]};
+ if (defined($aa)) {
+ my @aaa = split(",", $aa);
+ for (@aaa) {
+ $hit = 1 if ($_ eq $s[0]);
+ }
+ }
+ } else {
+ $hit = defined($h{$t[0],$t[1]})? 1 : 0;
+ }
+ push(@a, [$t[5], ($t[4] eq '.' || $t[4] eq $t[3])? 0 : 1, $ts{$t[3].$s[0]}? 1 : 0, $hit]);
}
push(@a, [-1, 0, 0, 0]); # end marker
die("[qstats] No SNP data!\n") if (@a == 0);
my $next = $opts{s};
my $last = $a[0];
my @c = (0, 0, 0, 0);
+ my @lc;
+ $lc[1] = $lc[2] = 0;
for my $p (@a) {
if ($p->[0] == -1 || ($p->[0] != $last && $c[0]/@a > $next)) {
my @x;
$x[0] = sprintf("%.4f", $c[1]-$c[2]? $c[2] / ($c[1] - $c[2]) : 100);
$x[1] = sprintf("%.4f", $hsize? $c[3] / $hsize : 0);
$x[2] = sprintf("%.4f", $c[3] / $c[1]);
+ my $a = $c[1] - $lc[1];
+ my $b = $c[2] - $lc[2];
+ $x[3] = sprintf("%.4f", $a-$b? $b / ($a-$b) : 100);
print join("\t", $last, @c, @x), "\n";
$next = $c[0]/@a + $opts{s};
+ $lc[1] = $c[1]; $lc[2] = $c[2];
}
++$c[0]; $c[1] += $p->[1]; $c[2] += $p->[2]; $c[3] += $p->[3];
$last = $p->[0];
}
}
+sub ucscsnp2vcf {
+ die("Usage: vcfutils.pl <in.ucsc.snp>\n") if (@ARGV == 0 && -t STDIN);
+ print "##fileformat=VCFv4.0\n";
+ print join("\t", "#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO"), "\n";
+ while (<>) {
+ my @t = split("\t");
+ my $indel = ($t[9] =~ /^[ACGT](\/[ACGT])+$/)? 0 : 1;
+ my $pos = $t[2] + 1;
+ my @alt;
+ push(@alt, $t[7]);
+ if ($t[6] eq '-') {
+ $t[9] = reverse($t[9]);
+ $t[9] =~ tr/ACGTRYMKWSNacgtrymkwsn/TGCAYRKMWSNtgcayrkmwsn/;
+ }
+ my @a = split("/", $t[9]);
+ for (@a) {
+ push(@alt, $_) if ($_ ne $alt[0]);
+ }
+ if ($indel) {
+ --$pos;
+ for (0 .. $#alt) {
+ $alt[$_] =~ tr/-//d;
+ $alt[$_] = "N$alt[$_]";
+ }
+ }
+ my $ref = shift(@alt);
+ my $af = $t[13] > 0? ";AF=$t[13]" : '';
+ my $valid = ($t[12] eq 'unknown')? '' : ";valid=$t[12]";
+ my $info = "molType=$t[10];class=$t[11]$valid$af";
+ print join("\t", $t[1], $pos, $t[4], $ref, join(",", @alt), 0, '.', $info), "\n";
+ }
+}
+
sub hapmap2vcf {
die("Usage: vcfutils.pl <in.ucsc.snp> <in.hapmap>\n") if (@ARGV == 0);
my $fn = shift(@ARGV);
qstats SNP stats stratified by QUAL
varFilter filtering short variants
hapmap2vcf convert the hapmap format to VCF
+ ucscsnp2vcf convert UCSC SNP SQL dump to VCF
\n/);
}