Recalculate coding TMB with updated Lancet BED file

analyses/snv-callers/run_caller_consensus_analysis.sh

@@ -15,11 +15,11 @@ dbfile=scratch/snv_db.sqlite
 consensus_file=analyses/snv-callers/results/consensus/pbta-snv-consensus-mutation.maf.tsv
 
 # BED and GTF file paths
-exon_file=scratch/gencode.v27.primary_assembly.annotation.bed
+cds_file=scratch/gencode.v27.primary_assembly.annotation.bed
 all_mut_wgs_bed=scratch/intersect_strelka_mutect_WGS.bed
 all_mut_wxs_bed=data/WXS.hg38.100bp_padded.bed
-coding_wgs_bed=scratch/intersect_exon_lancet_strelka_mutect_WGS.bed
-coding_wxs_bed=scratch/intersect_exon_WXS.bed
+coding_wgs_bed=scratch/intersect_cds_lancet_strelka_mutect_WGS.bed
+coding_wxs_bed=scratch/intersect_cds_lancet_WXS.bed
 
 # Set a default for the VAF filter if none is specified
 vaf_cutoff=${OPENPBTA_VAF_CUTOFF:-0}
@@ -60,20 +60,21 @@ bedtools intersect \
 gunzip -c data/gencode.v27.primary_assembly.annotation.gtf.gz \
   | awk '$3 ~ /CDS/' \
   | convert2bed --do-not-sort --input=gtf - \
-  > $exon_file
+  > $cds_file
 
 # Make WGS coding BED file
 bedtools intersect \
   -a data/WGS.hg38.strelka2.unpadded.bed \
   -b data/WGS.hg38.mutect2.vardict.unpadded.bed \
   data/WGS.hg38.lancet.300bp_padded.bed \
-  $exon_file \
+  $cds_file \
   > $coding_wgs_bed
 
 # Make WXS coding BED file
 bedtools intersect \
   -a data/WXS.hg38.100bp_padded.bed  \
-  -b $exon_file 
+  -b data/WXS.hg38.lancet.400bp_padded.bed \
+  $cds_file \
   > $coding_wxs_bed
 
 ######################### Calculate consensus TMB ##############################

analyses/snv-callers/scripts/03-calculate_tmb.R

@@ -29,8 +29,8 @@
 # --metadata data/pbta-histologies.tsv \
 # --all_bed_wgs scratch/intersect_strelka_mutect_WGS.bed \
 # --all_bed_wxs data/WXS.hg38.100bp_padded.bed \
-# --coding_bed_wgs scratch/intersect_exon_lancet_strelka_mutect_WGS.bed \
-# --coding_bed_wxs scratch/intersect_exon_WXS.bed \
+# --coding_bed_wgs scratch/intersect_cds_lancet_strelka_mutect_WGS.bed \
+# --coding_bed_wxs scratch/intersect_cds_WXS.bed \
 # --overwrite
 
 ################################ Initial Set Up ################################
@@ -205,7 +205,7 @@ if (file.exists(tmb_coding_file) && !opt$overwrite) {
   # Calculate coding only TMBs and write to file
   tmb_coding_df <- calculate_tmb(maf_df,
     bed_wgs = opt$coding_bed_wgs,
-    bed_wxs = opt$coding_bed_wgs
+    bed_wxs = opt$coding_bed_wxs
   )
   readr::write_tsv(tmb_coding_df, tmb_coding_file)
 

analyses/tp53_nf1_score/run_classifier.sh

@@ -15,7 +15,7 @@ POLYA_PLOT=${OPENPBTA_POLYAPLOT:-1}
 data_dir="data"
 scratch_dir="scratch"
 # cds gencode bed file  
-exon_file="${scratch_dir}/gencode.v27.primary_assembly.annotation.bed"
+cds_file="${scratch_dir}/gencode.v27.primary_assembly.annotation.bed"
 consensus_file="${data_dir}/pbta-snv-consensus-mutation.maf.tsv.gz"
 clinical_file="${data_dir}/pbta-histologies.tsv"
 analysis_dir="analyses/tp53_nf1_score"
@@ -25,14 +25,14 @@ analysis_dir="analyses/tp53_nf1_score"
 gunzip -c ${data_dir}/gencode.v27.primary_assembly.annotation.gtf.gz \
   | awk '$3 ~ /CDS/' \
   | convert2bed --do-not-sort --input=gtf - \
-  > $exon_file
+  > $cds_file
 
 # Prep the SNV consensus data for evaluation downstream
 Rscript --vanilla ${analysis_dir}/00-tp53-nf1-alterations.R \
   --snvConsensus ${consensus_file} \
   --clinicalFile ${clinical_file} \
   --outputFolder ${analysis_dir}/results \
-  --gencode ${exon_file}
+  --gencode ${cds_file}
 
 # expression files for prediction
 collapsed_stranded="pbta-gene-expression-rsem-fpkm-collapsed.stranded.rds"