Merge pull request #193 from sigven/docker_unwrap

Remove docker command wrappers; CPSR refactor

pcgr/pcgr_vars.py

@@ -2,6 +2,14 @@
 
 from pcgr._version import __version__
 
+PCGR_VERSION = __version__
+DB_VERSION = '20220203'
+VEP_VERSION = '105'
+GENCODE_VERSION = '39'
+NCBI_BUILD_MAF = 'GRCh38'
+VEP_ASSEMBLY = 'GRCh38'
+MAX_VARIANTS_FOR_REPORT = 500_000
+
 tsites = {
     0: 'Any',
     1: 'Adrenal Gland',
@@ -37,12 +45,51 @@
 }
 
 tumor_sites = '\n'.join([f'{k} = {tsites[k]}' for k in tsites]) # for displaying in help
-PCGR_VERSION = __version__
-DB_VERSION = '20220203'
-VEP_VERSION = '105'
-GENCODE_VERSION = '39'
-NCBI_BUILD_MAF = 'GRCh38'
-VEP_ASSEMBLY = 'GRCh38'
-MAX_VARIANTS_FOR_REPORT = 500000
-DOCKER_IMAGE_VERSION = f'sigven/pcgr:{PCGR_VERSION}'
 
+GE_panels = {
+      0: "CPSR exploratory cancer predisposition panel (n = 433, GEP / TCGA Germline Study / Cancer Gene Census / Other)",
+      1: "Adult solid tumours cancer susceptibility (GEP)",
+      2: "Adult solid tumours for rare disease (GEP)",
+      3: "Bladder cancer pertinent cancer susceptibility (GEP)",
+      4: "Brain cancer pertinent cancer susceptibility (GEP)",
+      5: "Breast cancer pertinent cancer susceptibility (GEP)",
+      6: "Childhood solid tumours cancer susceptibility (GEP)",
+      7: "Colorectal cancer pertinent cancer susceptibility (GEP)",
+      8: "Endometrial cancer pertinent cancer susceptibility (GEP)",
+      9: "Familial Tumours Syndromes of the central & peripheral Nervous system (GEP)",
+      10: "Familial breast cancer (GEP)",
+      11: "Familial melanoma (GEP)",
+      12: "Familial prostate cancer (GEP)",
+      13: "Familial rhabdomyosarcoma (GEP)",
+      14: "GI tract tumours (GEP)",
+      15: "Genodermatoses with malignancies (GEP)",
+      16: "Haematological malignancies cancer susceptibility (GEP)",
+      17: "Haematological malignancies for rare disease (GEP)",
+      18: "Head and neck cancer pertinent cancer susceptibility (GEP)",
+      19: "Inherited MMR deficiency (Lynch Syndrome) - GEP",
+      20: "Inherited non-medullary thyroid cancer (GEP)",
+      21: "Inherited ovarian cancer (without breast cancer) (GEP)",
+      22: "Inherited pancreatic cancer (GEP)",
+      23: "Inherited polyposis (GEP)",
+      24: "Inherited predisposition to acute myeloid leukaemia (AML) (GEP)",
+      25: "Inherited predisposition to GIST (GEP)",
+      26: "Inherited renal cancer (GEP)",
+      27: "Inherited phaeochromocytoma and paraganglioma (GEP)",
+      28: "Melanoma pertinent cancer susceptibility (GEP)",
+      29: "Multiple endocrine tumours (GEP)",
+      30: "Multiple monogenic benign skin tumours (GEP)",
+      31: "Neuroendocrine cancer pertinent cancer susceptibility (GEP)",
+      32: "Neurofibromatosis Type 1 (GEP)",
+      33: "Ovarian cancer pertinent cancer susceptibility (GEP)",
+      34: "Parathyroid Cancer (GEP)",
+      35: "Prostate cancer pertinent cancer susceptibility (GEP)",
+      36: "Renal cancer pertinent cancer susceptibility (GEP)",
+      37: "Rhabdoid tumour predisposition (GEP)",
+      38: "Sarcoma cancer susceptibility (GEP)",
+      39: "Sarcoma susceptbility (GEP)",
+      40: "Thyroid cancer pertinent cancer susceptibility (GEP)",
+      41: "Tumour predisposition - childhood onset (GEP)",
+      42: "Upper gastrointestinal cancer pertinent cancer susceptibility (GEP)"
+}
+
+panels = '\n'.join([f'{k} = {GE_panels[k]}' for k in GE_panels]) # for displaying in help

pcgr/utils.py

@@ -6,23 +6,6 @@
 import os
 import platform
 
-def get_docker_user_id(docker_user_id):
-    logger = getlogger('pcgr-get-OS')
-    uid = ''
-    if docker_user_id:
-        uid = docker_user_id
-    elif platform.system() == 'Linux' or platform.system() == 'Darwin' or sys.platform == 'darwin' or sys.platform == 'linux2' or sys.platform == 'linux':
-        uid = os.getuid()
-    else:
-        if platform.system() == 'Windows' or sys.platform == 'win32' or sys.platform == 'cygwin':
-            uid = getpass.getuser()
-
-    if uid == '':
-        warn_msg = (f'Was not able to get user id/username for logged-in user on the underlying platform '
-                    f'(platform.system(): {platform.system()},  sys.platform: {sys.platform}, now running PCGR as root')
-        logger.warning(warn_msg)
-        uid = 'root'
-    return uid
 
 def getlogger(logger_name):
     logger = logging.getLogger(logger_name)
@@ -60,25 +43,25 @@ def check_subprocess(logger, command, debug):
         print(e.output.decode())
         exit(0)
 
-def script_path(env, bin_script, docker_run):
+def script_path(env, bin_script):
     """Returns e.g. /path/conda/envs/{env}/{bin_script}
     """
-    prefix = conda_env_path(env, docker_run)
+    prefix = conda_env_path(env)
     return os.path.join(prefix, bin_script)
 
-def conda_env_path(env, docker_run):
+def conda_env_path(env):
     """Construct absolute path to a conda env
     using the current activated env as a prefix.
     e.g. /path/to/conda/envs/{env}
     """
-    if docker_run:
-        env_path = f'/opt/mambaforge/envs/{env}'
-    else:
-        cp = os.path.normpath(os.environ.get('CONDA_PREFIX'))  # /path/to/conda/envs/FOO
-        env_dir = os.path.dirname(cp)                          # /path/to/conda/envs
-        env_path = os.path.join(env_dir, env)                  # /path/to/conda/envs/{env}
+    cp = os.path.normpath(os.environ.get('CONDA_PREFIX'))  # /path/to/conda/envs/FOO
+    env_dir = os.path.dirname(cp)                          # /path/to/conda/envs
+    env_path = os.path.join(env_dir, env)                  # /path/to/conda/envs/{env}
     return env_path
 
+def get_loftee_dir():
+    return script_path("pcgr", "share/loftee")
+
 def get_pcgr_bin():
     """Return abs path to e.g. conda/env/pcgr/bin
     """
@@ -101,3 +84,17 @@ def get_perl_exports():
     perl_path_parent = os.path.dirname(perl_path) # /conda/env/pcgr/bin
     out = f"unset PERL5LIB && export PATH={perl_path_parent}:\"$PATH\""
     return out
+
+def is_integer(n):
+    try:
+        float(n)
+    except ValueError:
+        return False
+    else:
+        return float(n).is_integer()
+
+def get_cpsr_version():
+    # use pcgrr's Rscript to grab cpsr's R pkg version
+    rscript = script_path("pcgrr", "bin/Rscript")
+    v_cmd = f"{rscript} -e 'x <- paste0(\"cpsr \", as.character(packageVersion(\"cpsr\"))); cat(x, \"\n\")'"
+    return subprocess.check_output(v_cmd, shell=True).decode("utf-8")

pcgrr/vignettes/output.Rmd

@@ -320,8 +320,8 @@ A VCF file containing annotated, somatic calls (single nucleotide variants and i
 | `COSMIC_MUTATION_ID` | Mutation identifier in [Catalog of somatic mutations in cancer](http://cancer.sanger.ac.uk/cancergenome/projects/cosmic/) database, as provided by VEP |
 | `TCGA_PANCANCER_COUNT` | Raw variant count across all TCGA tumor types |
 | `TCGA_FREQUENCY` | Frequency of variant across TCGA tumor types. Format: `tumortype| percent affected|affected cases|total cases` |
-| `ICGC_PCAWG_OCCURRENCE` | Mutation occurrence in [ICGC|PCAWG](http://docs.icgc.org/pcawg/). By project: `project_code|affected_donors|tested_donors|frequency` |
-| `ICGC_PCAWG_AFFECTED_DONORS` | Number of donors with the current mutation in [ICGC|PCAWG](http://docs.icgc.org/pcawg/) |
+| `ICGC_PCAWG_OCCURRENCE` | Mutation occurrence in [ICGC-PCAWG](http://docs.icgc.org/pcawg/). By project: `project_code|tumor_type|affected_donors|tested_donors|frequency` |
+| `ICGC_PCAWG_AFFECTED_DONORS` (**?**) | Number of donors with the current mutation in [ICGC-PCAWG](http://docs.icgc.org/pcawg/) |
 
 ##### _Clinical associations_