需求处理

codes/chemo.py

@@ -104,26 +104,39 @@ class ChemoRun:
                 alt = record.alts[0]
                 # gt = '/'.join(list(map(str, sorted(record.samples.get(record.samples.keys()[0]).get('GT')))))
                 freq = record.samples.get(record.samples.keys()[-1]).get('AF')[0]
+                depth = record.samples.get(record.samples.keys()[-1]).get('DP')
                 if freq > 0.9:
                     gt = '1/1'
                 elif 0.9 >= freq > 0.1:
                     gt = '0/1'
                 else:
                     gt = '0/0'
-                fliter = pd.concat([fliter, drug_rsid_data[
+                match_drug_rsid_data = drug_rsid_data[
                     (drug_rsid_data['chr'] == chrom) &
                     (drug_rsid_data['end'] == end) &
                     (drug_rsid_data['ref'] == ref) &
                     (drug_rsid_data['alt'] == alt) &
                     (drug_rsid_data['genotype'] == gt)
-                    ]])
+                    ]
+                match_drug_rsid_data = match_drug_rsid_data.reset_index()
+                match_drug_rsid_data['chr'] = chrom
+                match_drug_rsid_data['pos'] = end
+                match_drug_rsid_data['freq'] = freq
+                match_drug_rsid_data['depth'] = depth
+                fliter = pd.concat([fliter, match_drug_rsid_data])
 
             if fliter.empty:
-                fliter = pd.concat([fliter, drug_rsid_data[
+                match_drug_rsid_data = drug_rsid_data[
                     (drug_rsid_data['chr'] == chrom) &
                     (drug_rsid_data['end'] == end) &
                     (drug_rsid_data['genotype'] == '0/0')
-                    ]])
+                    ]
+                match_drug_rsid_data = match_drug_rsid_data.reset_index()
+                match_drug_rsid_data['chr'] = chrom
+                match_drug_rsid_data['pos'] = end
+                match_drug_rsid_data['freq'] = '.'
+                match_drug_rsid_data['depth'] = '.'
+                fliter = pd.concat([fliter, match_drug_rsid_data])
 
             if fliter.empty:
                 raise UserWarning(
@@ -136,7 +149,8 @@ class ChemoRun:
         fliterdata.to_csv(respath, sep='\t', index=False)
 
         # 分类汇总 同位点，药物合并 drug.infos.txt
-        drugrsid = fliterdata[['drugname', 'genename', 'rsid', 'result', 'level', 'tips', 'drugsort']]
+        drugrsid = fliterdata[
+            ['drugname', 'genename', 'rsid', 'result', 'level', 'tips', 'drugsort', 'chr', 'pos', 'freq', 'depth']]
         drugrsid = drugrsid.drop_duplicates()
         resdrugrsid = drugrsid.groupby(['drugname', 'genename', 'rsid', 'result', 'level', 'drugsort'])['tips'].agg(
             ','.join).reset_index()

codes/filter_snpindel.pl

@@ -30,7 +30,8 @@ chomp($head);
 
 my @columns = split("\t", $head);
 my $new_head = join("\t", "Validated", "ClinicalSign", @columns[0 .. 6],
-    "Freq", @columns[7 .. 20, 23, 28, 32, 50, 56, 62, 101, 102], "Oncogenic", "Mutation_Effect", "genetag", "process");
+    "Freq", @columns[7 .. 20, 23, 28, 32, 50, 56, 62, 101, 102], "Oncogenic", "Mutation_Effect", "is_oncogene",
+    "is_tumor_suppressor_gene", "genetag", "process");
 
 if (!($pipeline eq 'somatic' || $pipeline eq 'tmb' || $pipeline eq 'hotspot' || $pipeline eq 'germline')) {
     die "useage: pipeline must be 'somatic' or 'germline' or 'hotspot or tmb'";
@@ -50,7 +51,9 @@ my @hhr2 = @$hhr2_ref;
 my @promoter = @$promoter_ref;
 
 my %transcript = transcript();
-my %oncogenic = get_oncogenic();
+my ($oncogenic, $is_oncogene) = get_oncogenic();
+my %oncogenic = %$oncogenic;
+my %is_oncogene = %$is_oncogene;
 
 while (<IN>) {
     chomp;
@@ -144,7 +147,7 @@ while (<IN>) {
     }
     elsif ($line[9] eq '.') {
         # splicing 位点
-        if ($line[5] =~ /splicing/) {
+        if (($line[5] =~ /splicing/) or ($pipeline eq 'hotspot')) {
             my @hgvs = split(/;/, $line[7]);
             my $hgvs = $hgvs[0];
             my $transcript_gene;
@@ -166,6 +169,11 @@ while (<IN>) {
                 $hgvs =~ s/exon(\d+)/intron$intron;exon$exon/;
                 $line[9] = join(":", ($gene, $hgvs));
             }
+            # 不是前面2种情况，hotspot强制转换hgvs
+            elsif ($pipeline eq 'hotspot') {
+                print "$hgvs\n";
+                $line[9] = join(":", ($gene, $hgvs));
+            }
             else {
                 push @reason, 'not_need_spl';
             }
@@ -177,6 +185,9 @@ while (<IN>) {
 
     }
     else {
+        if ($line[8] eq 'intron') {
+            push @reason, 'not_need_spl_inron';
+        }
         my @hgvs = split(/,/, $line[9]);
         my $hgvs = $hgvs[0];
         my $transcript_gene;
@@ -214,6 +225,13 @@ while (<IN>) {
     if ((grep {$_ eq $gene} @promoter) and ($pipeline eq 'somatic') and ($gene eq 'TERT')
         and ($line[1] eq '1295228' and $line[4] eq 'A') or ($line[1] eq '1295250' and $line[4] eq 'A')) {
         @reason = ();
+        if ($line[1] eq '1295228') {
+            $line[9] = 'TERT:NM_198253:/:c.-124C>T (C228T)';
+        }
+        else {
+            $line[9] = 'TERT:NM_198253:/:c.-146C>T (C250T)';
+        }
+        $line[8] = 'promoter';
     }
 
     if (@reason) {
@@ -221,17 +239,27 @@ while (<IN>) {
         next;
     }
 
-    my ($oncogenic_col, $mut_effect_col);
+    my ($oncogenic_col, $mut_effect_col, $is_oncogene_gene, $is_tumor_suppressor_gene);
     my $get_key = "$gene\_$protein";
     if (exists $oncogenic{lc $get_key}) {
         my @get_values = split('&&', $oncogenic{lc $get_key});
         $oncogenic_col = $get_values[0];
         $mut_effect_col = $get_values[1];
+
     }
     else {
         $oncogenic_col = '.';
         $mut_effect_col = '.';
     }
+    if (exists $is_oncogene{lc $gene}) {
+        my @get_values = split('&&', $is_oncogene{lc $gene});
+        $is_oncogene_gene = $get_values[0];
+        $is_tumor_suppressor_gene = $get_values[1];
+    }
+    else {
+        $is_oncogene_gene = '.';
+        $is_tumor_suppressor_gene = '.';
+    }
 
     my $clisig;
     if ($line[16] =~ /Likely_pathogenic|drug/i) {
@@ -266,7 +294,9 @@ while (<IN>) {
 
     $line[6] = $gene;
     my $genetag = join(";", @genetags);
-    my $new_line = join("\t", $validated, $clisig, @line[0 .. 6], $freq, @line[7 .. 20, 23, 28, 32, 50, 56, 62, 101, 102], $oncogenic_col, $mut_effect_col, $genetag, $pipeline);
+    my $new_line = join("\t", $validated, $clisig,
+        @line[0 .. 6], $freq, @line[7 .. 20, 23, 28, 32, 50, 56, 62, 101, 102],
+        $oncogenic_col, $mut_effect_col, $is_oncogene_gene, $is_tumor_suppressor_gene, $genetag, $pipeline);
     print OUT "$new_line\n";
     print TAG_OUT "PASS\t", join("\t", @line), "\n";
 
@@ -339,13 +369,16 @@ sub transcript {
 # oncokb snv_indel 临床意义定义
 sub get_oncogenic {
     my %sig;
+    my %sig_gene;
     open SNV_INDEL, "$database_path/snv_indel_mutation.csv";
     <SNV_INDEL>;
     while (<SNV_INDEL>) {
         chomp;
+        $_ =~ s/\r//g;
         my @line = split(",");
         my $key = join("_", @line[0, 1]);
-        $sig{lc $key} = join("&&", @line[2, 3]);
+        $sig{lc $key} = join("&&", @line[2, 3, 7, 8]);
+        $sig_gene{lc $line[0]} = join("&&", @line[7, 8]);
     }
-    return %sig;
+    return (\%sig, \%sig_gene);
 }

codes/postprocess.py

@@ -263,6 +263,7 @@ class PostProcess:
             filter_sum_df = filter_sum_df[cols]
 
         filter_sum_df = filter_sum_df.fillna('.')
+        filter_sum_df = filter_sum_df.sort_values(by='AMP_mut_level')
         filter_sum_res = filter_sum_df.to_dict('records')
 
         self.sheet['target_mut'] = filter_sum_res