PoC GTEx tissue scores

.gitignore

@@ -17,6 +17,7 @@
 !/resources/README.txt
 !/resources/decision_tree_*.json
 /resources/vep/cache
+resources/GTEx/GTEx_Analysis_2017-06-05_v8_RSEMv1.3.0_transcript_tpm.gct
 !/resources/vep/plugins
 !/resources/*.chain
 !/resources/*.chain.gz
@@ -33,4 +34,4 @@ nextflow-*-all
 /vip
 
 # mkdocs
-/site
+/site

config/nxf_vcf.config

@@ -60,6 +60,8 @@ params {
       vep_plugin_inheritance = "${projectDir}/resources/inheritance_20240115.tsv"
       vep_plugin_vkgl_mode = 1
 
+      vep_plugin_gtex = "${projectDir}/resources/GTEx/GTEx_Analysis_2017-06-05_v8_RSEMv1.3.0_transcript_tpm.gct"
+
       GRCh38 {
         capice_model = "${projectDir}/resources/GRCh38/capice_model_v5.1.2-v2.ubj"
         expansionhunter_variant_catalog = "${projectDir}/resources/GRCh38/expansionhunter_variant_catalog.json"
@@ -83,7 +85,8 @@ params {
 			metadata = "${projectDir}/resources/field_metadata.json"
 
       GRCh38 {
-        decision_tree = "${projectDir}/resources/decision_tree_GRCh38.json"
+        //decision_tree = "${projectDir}/resources/decision_tree_GRCh38.json"
+        decision_tree = "${projectDir}/resources/PoC/decision_tree_GRCh38_Tissues.json"
       }
     }
 
@@ -109,7 +112,7 @@ params {
       include_crams = true
       max_records = ""
       max_samples = ""
-      template = ""
+      template = "${projectDir}/resources/PoC/index.html"
 			metadata = "${projectDir}/resources/field_metadata.json"
 
       GRCh38 {

install.sh

@@ -144,6 +144,9 @@ download_files() {
   for ((i = 0; i < ${#urls[@]}; i += 2)); do
     download_file "${base_url}" "${urls[i+1]}" "${urls[i+0]}" "${output_dir}" "${validate}"
   done
+  download_file "https://ftp.ensembl.org/pub/release-111/variation/indexed_vep_cache" "homo_sapiens_vep_111_GRCh38.tar.gz" "FIXME" "${output_dir}/resources/vep/cache/" "false"
+  download_file "https://storage.googleapis.com/adult-gtex/bulk-gex/v8/rna-seq" "GTEx_Analysis_2017-06-05_v8_RSEMv1.3.0_transcript_tpm.gct.gz" "FIXME" "${output_dir}/resources/GTEx" "false"
+
 }
 
 extract_files() {
@@ -162,7 +165,16 @@ extract_files() {
     echo -e "extracting ${vep_gz} ..."
     tar -xzf "${vep_gz}" -C "${vep_dir}"
   fi
+  if [ ! -d "${vep_dir}/homo_sapiens/111_GRCh38" ]; then
+    local -r vep_gz="${vep_dir}/homo_sapiens_vep_111_GRCh38.tar.gz"
+    echo -e "extracting ${vep_gz} ..."
+    tar -xzf "${vep_gz}" -C "${vep_dir}"
+  fi
 
+  if [ ! -f "${output_dir}/resources/GTEx/GTEx_Analysis_2017-06-05_v8_RSEMv1.3.0_transcript_tpm.gct" ]; then
+    gunzip ${output_dir}/resources/GTEx/GTEx_Analysis_2017-06-05_v8_RSEMv1.3.0_transcript_tpm.gct.gz
+  fi
+
   local -r annotsv_dir="${output_dir}/resources/annotsv/v3.3.6"
 
   local -r annotsv_human_dir="${annotsv_dir}/Annotations_Human"

modules/sample_sheet.nf

@@ -42,6 +42,10 @@ def parseCommonSampleSheet(csvFilename, additionalCols) {
       list: true,
       regex: /HP:\d{7}/
     ],
+    tissues: [
+      type: "string",
+      list: true
+    ],
     sequencing_method: [
       type: "string",
       default: { 'WGS' },

modules/vcf/annotate.nf

@@ -1,4 +1,4 @@
-include { basename; areProbandHpoIdsIndentical } from './utils'
+include { basename; areProbandHpoIdsIndentical; getTissues } from './utils'
 
 process annotate {
   label 'vcf_annotate'
@@ -32,8 +32,10 @@ process annotate {
     capiceModelPath = params.vcf.annotate[assembly].capice_model
     alphScorePath = params.vcf.annotate[assembly].vep_plugin_alphscore
     strangerCatalog = params.vcf.annotate[assembly].stranger_catalog
+    gtexFile = params.vcf.annotate.vep_plugin_gtex
 
     areProbandHpoIdsIndentical = areProbandHpoIdsIndentical(meta.project.samples)
+    tissues = getTissues(meta.project.samples)
     gadoScores = meta.gado != null ? meta.gado : ""
 
     template 'annotate.sh'

modules/vcf/classify.nf

@@ -1,4 +1,4 @@
-include { basename } from './utils'
+include { basename; getTissues } from './utils'
 
 process classify {
   label 'vcf_classify'
@@ -18,7 +18,8 @@ process classify {
     metadata = params.vcf.classify.metadata
     decisionTree = params.vcf.classify[meta.project.assembly].decision_tree
     annotatePath = params.vcf.classify.annotate_path
-
+    tissues = getTissues(meta.project.samples)
+
     template 'classify.sh'
 
   stub:

modules/vcf/templates/annotate.sh

@@ -139,7 +139,7 @@ vep() {
   args+=("--dir_cache" "!{params.vcf.annotate.vep_cache_dir}")
   args+=("--species" "homo_sapiens")
   args+=("--assembly" "!{assembly}")
-  args+=("--refseq")
+  #args+=("--refseq")
   args+=("--exclude_predicted")
   args+=("--use_given_ref")
   args+=("--symbol")
@@ -160,20 +160,24 @@ vep() {
   args+=("--dir_plugins" "!{params.vcf.annotate.vep_plugin_dir}")
   args+=("--plugin" "Grantham")
   args+=("--plugin" "SpliceAI,snv=!{vepPluginSpliceAiSnvPath},indel=!{vepPluginSpliceAiIndelPath}")
-  args+=("--plugin" "Capice,${capiceOutputPath}")
+  args+=("--plugin" "Capice,${capiceOutputPath},!{params.vcf.annotate.ensembl_gene_mapping}")
   args+=("--plugin" "UTRannotator,!{vepPluginUtrAnnotatorPath}")
   args+=("--custom" "!{vepCustomPhyloPPath},phyloP,bigwig,exact,0")
   args+=("--safe")
 
   if [ -n "!{hpoIds}" ]; then
-    args+=("--plugin" "Hpo,!{params.vcf.annotate.vep_plugin_hpo},!{hpoIds.replace(',', ';')}")
+    args+=("--plugin" "Hpo,!{params.vcf.annotate.vep_plugin_hpo},!{hpoIds.replace(',', ';')},!{params.vcf.annotate.ensembl_gene_mapping}")
   fi
   if [ -n "!{gadoScores}" ]; then
     args+=("--plugin" "GADO,!{gadoScores},!{params.vcf.annotate.ensembl_gene_mapping}")
   fi
-  args+=("--plugin" "Inheritance,!{params.vcf.annotate.vep_plugin_inheritance}")
+  if [ -n "!{gtexFile}" ]; then
+    echo !{tissues}
+    args+=("--plugin" "GTEx,!{gtexFile},!{tissues.replace(',', ';')}")
+  fi
+  args+=("--plugin" "Inheritance,!{params.vcf.annotate.vep_plugin_inheritance},,!{params.vcf.annotate.ensembl_gene_mapping}")
   if [ -n "!{vepPluginVkglPath}" ] && [ -n "!{params.vcf.annotate.vep_plugin_vkgl_mode}" ]; then
-    args+=("--plugin" "VKGL,!{vepPluginVkglPath},!{params.vcf.annotate.vep_plugin_vkgl_mode}")
+    args+=("--plugin" "VKGL,!{vepPluginVkglPath},!{params.vcf.annotate.vep_plugin_vkgl_mode},!{params.vcf.annotate.ensembl_gene_mapping}")
   fi
   if [ -n "!{vepPluginGnomAdPath}" ]; then
     args+=("--plugin" "gnomAD,!{vepPluginGnomAdPath}")

modules/vcf/templates/classify.sh

@@ -9,7 +9,7 @@ classify () {
   args+=("-jar" "/opt/vcf-decision-tree/lib/vcf-decision-tree.jar")
   args+=("--input" "!{vcf}")
   args+=("--metadata" "!{metadata}")
-  args+=("--config" "!{decisionTree}")
+  args+=("--config" "decision_tree_updated.json")
   if [ !{annotatePath} -eq 1 ]; then
     args+=("--path")
   fi
@@ -42,7 +42,18 @@ insert_alt(){
   fi
 }
 
+write_tissue_file(){
+  echo !{tissues} | tr ',' '\n' > tissues.tsv
+}
+
+update_tree(){
+  tissuePath=$(realpath tissues.tsv)
+  sed "s|TISSUE_FILE_PATH|${tissuePath}|g" "!{decisionTree}" > decision_tree_updated.json
+}
+
 main () {
+  write_tissue_file
+  update_tree
   store_alt
   classify
   insert_alt

modules/vcf/utils.nf

@@ -73,6 +73,18 @@ def determineChunks(meta) {
   return chunks
 }
 
+def getTissues(samples) {
+  def tissueKeys = samples.collectMany { sample -> sample.tissues }.unique()
+  def allTissues = ["Adipose_Subcutaneous","Adipose_Visceral","AdrenalGland","Artery_Aorta","Artery_Coronary","Artery_Tibial","Bladder","Brain_Amygdala","Brain_Anteriorcingulatecortex","Brain_Caudate","Brain_CerebellarHemisphere","Brain_Cerebellum","Brain_Cortex","Brain_FrontalCortex","Brain_Hippocampus","Brain_Hypothalamus","Brain_Nucleusaccumbens","Brain_Putamen","Brain_Spinalcord","Brain_Substantianigra","Breast_MammaryTissue","Cells_Culturedfibroblasts","Cells_EBV_transformedlymphocytes","Cervix_Ectocervix","Cervix_Endocervix","Colon_Sigmoid","Colon_Transverse","Esophagus_GastroesophagealJunction","Esophagus_Mucosa","Esophagus_Muscularis","FallopianTube","Heart_AtrialAppendage","Heart_LeftVentricle","Kidney_Cortex","Kidney_Medulla","Liver","Lung","MinorSalivaryGland","Muscle_Skeletal","Nerve_Tibial","Ovary","Pancreas","Pituitary","Prostate","Skin_NotSunExposed","Skin_SunExposed","SmallIntestine_TerminalIleum","Spleen","Stomach","Testis","Thyroid","Uterus","Vagina","WholeBlood"] as String[]
+  def tissues = []
+  for(tissueKey in tissueKeys){
+    tissues.addAll(Arrays.stream(allTissues)
+          .filter(tissue -> tissue.matches(".*${tissueKey}.*"))
+          .toArray(size -> new String[size]));
+  }
+  return tissues.join(",")
+}
+
 def scatter(meta) {
     def chunks = determineChunks(meta)
     def index = 0