Add transcript export file #256

config/param.transcripts.json

@@ -7,7 +7,8 @@
  "from_column_format": [
  {
  "transcripts_column": "ANN",
- "transcripts_infos_column": "Feature_ID"
+ "transcripts_infos_column": "Feature_ID",
+ "column_clean": true
  }
  ],
  "from_columns_map": [
@@ -35,6 +36,9 @@
  "prioritization_config": "config/prioritization_transcripts_profiles.json",
  "pzprefix": "PZT",
  "prioritization_score_mode": "HOWARD"
+ },
+ "export": {
+ "output": "/tmp/output.transcripts.tsv"
  }
  }
 }

howard/objects/variants.py

@@ -2273,6 +2273,7 @@ def export_output(
  query=query,
  export_header=export_header,
  sample_list=sample_list,
+ table="transcripts",
  )
 
  # Remove
@@ -6622,6 +6623,14 @@ def get_config_default(self, name: str) -> dict:
  "function_name": "calculation_transcripts_prioritization",
  "function_params": [],
  },
+ "transcripts_export": {
+ "type": "python",
+ "name": "transcripts_export",
+ "description": "Export transcripts table/view as a file (using param.json)",
+ "available": True,
+ "function_name": "calculation_transcripts_export",
+ "function_params": [],
+ },
  },
  "prioritizations": {
  "default": {
@@ -9612,10 +9621,135 @@ def calculation_transcripts_prioritization(self) -> None:
  else:
  log.info("No Transcripts to process. Check param.json file configuration")
 
+ def calculation_transcripts_export(self) -> None:
+ """ """
+
+ # Create transcripts table
+ transcripts_table = self.create_transcript_view()
+
+ # Add info field
+ if transcripts_table:
+ self.transcripts_export(transcripts_table=transcripts_table)
+ else:
+ log.info("No Transcripts to process. Check param.json file configuration")
+
+
  ###############
  # Transcripts #
  ###############
 
+ def transcripts_export(
+ self, transcripts_table: str = None, param: dict = {}
+ ) -> bool:
+ """ """
+
+ log.debug("Start transcripts export...")
+
+ # Param
+ if not param:
+ param = self.get_param()
+
+ # Param export
+ param_transcript_export = param.get("transcripts", {}).get("export", {})
+
+ # Output file
+ transcripts_export_output = param_transcript_export.get("output", None)
+
+ if not param_transcript_export or not transcripts_export_output:
+ log.warning(f"No transcriipts export parameters defined!")
+ return False
+
+ # List of transcripts annotations
+ query_describe = f"""
+ SELECT column_name
+ FROM (
+ DESCRIBE SELECT * FROM {transcripts_table}
+ )
+ WHERE column_name NOT IN ('#CHROM', 'POS', 'REF', 'ALT', 'INFO')
+ """
+ transcripts_annotations_list = list(
+ self.get_query_to_df(query=query_describe)["column_name"]
+ )
+
+ # Create transcripts table for export
+ transcripts_table_export = f"{transcripts_table}_export_" + "".join(
+ random.choices(string.ascii_uppercase + string.digits, k=10)
+ )
+ query_create_transcripts_table_export = f"""
+ CREATE TABLE {transcripts_table_export} AS (SELECT "#CHROM", "POS", "REF", "ALT", '' AS 'INFO', {', '.join(transcripts_annotations_list)} FROM {transcripts_table})
+ """
+ self.execute_query(query=query_create_transcripts_table_export)
+
+ # Output file format
+ transcripts_export_output_format = get_file_format(
+ filename=transcripts_export_output
+ )
+
+ # Format VCF - construct INFO
+ if transcripts_export_output_format in ["vcf"]:
+
+ # Construct query update INFO and header
+ query_update_info = []
+ for field in transcripts_annotations_list:
+
+ # If field not in header
+ if field not in self.get_header_infos_list():
+
+ # Add PZ Transcript in header
+ self.get_header().infos[field] = vcf.parser._Info(
+ field,
+ ".",
+ "String",
+ f"Annotation '{field}' from transcript view",
+ "unknown",
+ "unknown",
+ 0,
+ )
+
+ # Add field as INFO/tag
+ query_update_info.append(
+ f"""
+ CASE
+ WHEN "{field}" IS NOT NULL
+ THEN concat('{field}=', "{field}", ';') 
+ ELSE '' 
+ END
+ """
+ )
+
+ # Query param
+ query_update_info_value = (
+ f""" concat('', {", ".join(query_update_info)}) """
+ )
+ query_export_columns = f""" "#CHROM", "POS", '.' AS 'ID', "REF", "ALT", '.' AS 'QUAL', '.' AS 'FILTER', "INFO" """
+
+ else:
+
+ # Query param
+ query_update_info_value = f""" NULL """
+ query_export_columns = f""" "#CHROM", "POS", "REF", "ALT", {', '.join(transcripts_annotations_list)} """
+
+ # Update query INFO column
+ query_update = f"""
+ UPDATE {transcripts_table_export}
+ SET INFO = {query_update_info_value}
+
+ """
+ self.execute_query(query=query_update)
+
+ # Export
+ self.export_output(
+ output_file=transcripts_export_output,
+ query=f""" SELECT {query_export_columns} FROM {transcripts_table_export} """,
+ )
+
+ # Drop transcripts export table
+ query_drop_transcripts_table_export = f"""
+ DROP TABLE {transcripts_table_export}
+ """
+ self.execute_query(query=query_drop_transcripts_table_export)
+
+
  def transcripts_prioritization(
  self, transcripts_table: str = None, param: dict = {}
  ) -> bool:

howard/tools/tools.py

@@ -1893,7 +1893,7 @@ def __call__(self, string):
  """ howard calculation --input=tests/data/example.ann.vcf.gz --output=/tmp/example.calculated.tsv --calculations='snpeff_hgvs,NOMEN' --hgvs_field=snpeff_hgvs --transcripts=tests/data/transcripts.tsv \n"""
  """ howard calculation --input=tests/data/example.vcf.gz --output=/tmp/example.calculated.tsv --calculations='TRIO' --trio_pedigree='sample1,sample2,sample4' \n"""
  """ howard calculation --input=tests/data/example.vcf.gz --output=/tmp/example.calculated.tsv --calculations='BARCODEFAMILY' --family_pedigree='sample1,sample2,sample4' \n"""
- """ howard calculation --input=tests/data/example.ann.transcripts.vcf.gz --output=/tmp/example.calculation.transcripts.tsv --param=config/param.transcripts.json --calculations='TRANSCRIPTS_ANNOTATIONS,TRANSCRIPTS_PRIORITIZATION' \n"""
+ """ howard calculation --input=tests/data/example.ann.transcripts.vcf.gz --output=/tmp/example.calculation.transcripts.tsv --param=config/param.transcripts.json --calculations='TRANSCRIPTS_ANNOTATIONS,TRANSCRIPTS_PRIORITIZATION,TRANSCRIPTS_EXPORT' \n"""
  """ howard calculation --input=tests/data/example.ann.vcf.gz --output=/tmp/example.ann.tsv --param=config/param.json \n"""
  """ howard calculation --show_calculations \n"""
  """ \n""",

tests/data/param.transcripts.json

@@ -0,0 +1,44 @@
+{
+ "transcripts": {
+ "table": "transcripts",
+ "transcripts_info_field_json": "transcripts_json",
+ "transcripts_info_field_format": "transcripts_ann",
+ "struct": {
+ "from_column_format": [
+ {
+ "transcripts_column": "ANN",
+ "transcripts_infos_column": "Feature_ID",
+ "column_clean": true
+ }
+ ],
+ "from_columns_map": [
+ {
+ "transcripts_column": "Ensembl_transcriptid",
+ "transcripts_infos_columns": [
+ "genename",
+ "Ensembl_geneid",
+ "LIST_S2_score",
+ "LIST_S2_pred"
+ ]
+ },
+ {
+ "transcripts_column": "Ensembl_transcriptid",
+ "transcripts_infos_columns": [
+ "genename",
+ "VARITY_R_score",
+ "Aloft_pred"
+ ]
+ }
+ ]
+ },
+ "prioritization": {
+ "profiles": ["transcripts"],
+ "prioritization_config": "config/prioritization_transcripts_profiles.json",
+ "pzprefix": "PZT",
+ "prioritization_score_mode": "HOWARD"
+ },
+ "export": {
+ "output": "/tmp/output.transcripts.tsv"
+ }
+ }
+}

tests/test_variants_transcripts.py

@@ -14,8 +14,9 @@
 from tempfile import TemporaryDirectory
 import pytest
 import vcf
+import os
 
-from howard.functions.commons import remove_if_exists
+from howard.functions.commons import remove_if_exists, get_file_format
 from howard.objects.variants import Variants
 from test_needed import tests_folder, tests_config, tests_data_folder
 
@@ -1710,3 +1711,97 @@ def test_transcripts_create_view_param_mapping(
  vcf.Reader(filename=output_vcf)
  except:
  assert False
+
+
+
+@pytest.mark.parametrize(
+ "output",
+ [
+ "output.vcf",
+ "output.vcf.gz",
+ "output.tsv",
+ "output.tsv.gz",
+ "output.parquet",
+ "output.json",
+ ],
+)
+def test_transcripts_create_view_export(output):
+ """ """
+
+ with TemporaryDirectory(dir=tests_folder) as tmp_dir:
+
+ tmp_dir = "/tmp"
+
+ # Init files
+ input_vcf = f"{tests_data_folder}/example.ann.transcripts.vcf.gz"
+
+ # Construct param dict
+ param_struct = {
+ "table": "transcripts",
+ "column_id": "transcript",
+ "transcripts_info_json": "transcripts_json",
+ "transcripts_info_field": "transcripts_json",
+ "transcript_id_remove_version": True,
+ "transcript_id_mapping_file": None,
+ "transcript_id_mapping_force": False,
+ "struct": {
+ "from_column_format": [
+ {
+ "transcripts_column": "ANN",
+ "transcripts_infos_column": "Feature_ID",
+ "column_clean": True,
+ },
+ ],
+ "from_columns_map": [
+ {
+ "transcripts_column": "Ensembl_transcriptid",
+ "transcripts_infos_columns": [
+ "genename",
+ "Ensembl_geneid",
+ "LIST_S2_score",
+ "LIST_S2_pred",
+ ],
+ "column_clean": False,
+ },
+ {
+ "transcripts_column": "Ensembl_transcriptid",
+ "transcripts_infos_columns": [
+ "genename",
+ "VARITY_R_score",
+ "Aloft_pred",
+ ],
+ "column_clean": False,
+ },
+ ],
+ },
+ "export": {"output": f"{tmp_dir}/{output}"},
+ }
+
+ # Param without prioritization
+ param_with_transcripts = {"transcripts": dict(param_struct)}
+
+ # Create object
+ variants = Variants(
+ conn=None, input=input_vcf, param=param_with_transcripts, load=True
+ )
+
+ # Create transcript view
+ transcripts_table = variants.create_transcript_view(
+ param=param_with_transcripts
+ )
+
+ # Export
+ try:
+ variants.transcripts_export(
+ transcripts_table=transcripts_table, param=param_with_transcripts
+ )
+ except:
+ assert False
+
+ assert os.path.isfile(f"{tmp_dir}/{output}")
+
+ if get_file_format(output) in ["vcf"]:
+ try:
+ vcf.Reader(filename=f"{tmp_dir}/{output}")
+ except:
+ assert False