Merge pull request #102 from eastgenomics/DI-286_Gemini_Rcode_matching

minor fixes for v2.0.1 (#102)

cnvreports.py

@@ -211,8 +211,9 @@ def run_cnvreports(
  panels.append(CI)
  else:
  clinical_indication = next(
- (key for key in CI2panels_dict.keys() if key == CI),
- None)
+ (key for key in CI2panels_dict.keys()
+ if key.split("_")[0] == CI.split("_")[0]), None
+ )
  if clinical_indication is None:
  # skip sample if CI not found
  skip_sample = True

general_functions.py

@@ -750,15 +750,20 @@ def parse_Gemini_manifest(manifest_file): # reports
  )
  sample_identifier = record[0] # X number
  clinical_indications = record[-1].split(",")
- CIs = list(set(
- [CI for CI in clinical_indications if CI.startswith("R") or CI.startswith("_")]
+ # expecting Test Codes to be comma-separated
+ # handle R###.# and C##.# test codes and _HGNC IDs
+ # strip whitespaces in case they get through somehow
+ test_codes = list(set(
+ [CI.strip(" ") for CI in clinical_indications if
+ re.search(r"^[RC][0-9]+\.[0-9]+", CI.strip(" ")) or
+ re.search(r"^_HGNC", CI.strip(" "))]
  ))
- # if sample is already assigned to a list of CIs, extend the list
+ # if sample is already assigned to a list of test_codes, extend the list
  if sample_identifier in data.keys():
- data[sample_identifier]["CIs"].append(CIs)
- # if sample has no CIs yet, save the list
+ data[sample_identifier]["test_codes"].extend(test_codes)
+ # if sample has no test_codes yet, save the list
  else:
- data[sample_identifier] = {"test_codes": CIs}
+ data[sample_identifier] = {"test_codes": test_codes}
 
  return data
 
@@ -898,7 +903,7 @@ def create_job_report_file(job__report_dict):
  )
  for sample, test in job__report_dict["invalid_tests"]:
  f.write(
- " * {}: {}".format(sample, test)
+ "{}\t{}\n".format(sample, test)
  )
 
  return job_report_file

reports.py

@@ -210,8 +210,9 @@ def run_reports(
  panels.append(CI)
  else:
  clinical_indication = next(
- (key for key in CI2panels_dict.keys() if key == CI),
- None)
+ (key for key in CI2panels_dict.keys()
+ if key.split("_")[0] == CI.split("_")[0]), None
+ )
  if clinical_indication is None:
  # skip sample if CI not found
  skip_sample = True