rcsb.workflow 0.45__tar.gz → 0.47__tar.gz

{rcsb_workflow-0.45 → rcsb_workflow-0.47}/HISTORY.txt

@@ -37,3 +37,6 @@
 21-May-2024 - V0.43 Fix pylinting
 20-Aug-2024 - V0.44 Add step for loading target cofactor data to MongoDB
 27-Aug-2024 - V0.45 Update usage of CARDTargetOntologyProvider
+22-Oct-2024 - V0.46 Update requirements and CI/CD (to use python 3.10)
+10-Dec-2024 - V0.47 Add support for 'max-seqs' flag in mmseqs database search;
+                    Update Azure pipelines to run on latest macOS and ubuntu version

{rcsb_workflow-0.45/rcsb.workflow.egg-info → rcsb_workflow-0.47}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rcsb.workflow
-Version: 0.45
+Version: 0.47
 Summary: RCSB Python data processing and ETL/ELT workflow entry points
 Home-page: https://github.com/rcsb/py-rcsb_workflow
 Author: John Westbrook
@@ -12,15 +12,16 @@ Classifier: Natural Language :: English
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: OpenEye-toolkits>=2020.2.2
+Requires-Dist: OpenEye-toolkits>=2024.1.1
 Requires-Dist: rcsb.utils.chem>=0.79
 Requires-Dist: rcsb.utils.seq>=0.82
 Requires-Dist: rcsb.utils.targets>=0.82
-Requires-Dist: rcsb.utils.seqalign>=0.29
-Requires-Dist: rcsb.exdb>=1.23
+Requires-Dist: rcsb.utils.seqalign>=0.31
+Requires-Dist: rcsb.exdb>=1.25
 Requires-Dist: rcsb.utils.io>=1.48
 Provides-Extra: dev
 Requires-Dist: check-manifest; extra == "dev"

{rcsb_workflow-0.45 → rcsb_workflow-0.47}/rcsb/workflow/chem/__init__.py

@@ -2,4 +2,4 @@ __docformat__ = "google en"
 __author__ = "John Westbrook"
 __email__ = "john.westbrook@rcsb.org"
 __license__ = "Apache 2.0"
-__version__ = "0.45"
+__version__ = "0.47"

{rcsb_workflow-0.45 → rcsb_workflow-0.47}/rcsb/workflow/targets/ProteinTargetSequenceExecutionWorkflow.py

@@ -8,6 +8,7 @@
 #  21-Mar-2023 Allow backing up Pharos-targets to stash, more __init__ improvement
 #   5-May-2023 Pass in fromDbPharos and reloadPharos parameters to exportFasta()
 #  12-Jun-2023 dwp Set useTaxonomy filter to False for CARD annotations
+#  10-Dec-2024 dwp Specify 'max-seqs' for mmseqs search to override default value
 ##
 """
 Execution workflow for protein target data ETL operations.
@@ -103,7 +104,9 @@ class ProteinTargetSequenceExecutionWorkflow(object):
         return ok
 
     def fetchProteinEntityData(self):
-        """Export RCSB protein entity sequence FASTA, taxonomy, and sequence details"""
+        """Export RCSB protein entity sequence FASTA, taxonomy, and sequence details
+        by fetching from 'pdbx_core_polymer_entity'.
+        """
         logger.info("Running fetchProteinEntityData...")
         ok = False
         try:
@@ -114,7 +117,13 @@ class ProteinTargetSequenceExecutionWorkflow(object):
         return ok
 
     def fetchChemicalReferenceMappingData(self):
-        """Export RCSB chemical reference identifier mapping details"""
+        """Export RCSB chemical reference identifier mapping details.
+
+        Fetch/prepare mapping of all chemical references from 'bird_chem_comp_core' (for DrugBank and ChEMBL)
+        (e.g., all CCs where {'rcsb_chem_comp_related.resource_name': 'DrugBank'}).
+
+            --> Creates file: CACHE/chemref-mapping/chemref-mapping-data.json
+        """
         logger.info("Running fetchChemicalReferenceMappingData...")
         ok = False
         try:
@@ -125,7 +134,14 @@ class ProteinTargetSequenceExecutionWorkflow(object):
         return ok
 
     def fetchLigandNeighborMappingData(self):
-        """Export RCSB ligand neighbor mapping details"""
+        """Export RCSB ligand neighbor mapping details.
+
+        Fetch/prepare mapping of all polymer_entities and their associated ligands (e.g., "2E1B_1": ["ZN"]).
+        This is done by extracting out all "rcsb_ligand_neighbors" from all polymer_entity_instances, and then
+        grouping them together on a per-entity basis.
+
+            --> Creates file: CACHE/ligand-neighbor-mapping/ligand-neighbor-mapping-data.json
+        """
         logger.info("Running fetchLigandNeighborMappingData...")
         ok = False
         try:
@@ -136,7 +152,19 @@ class ProteinTargetSequenceExecutionWorkflow(object):
         return ok
 
     def exportFasta(self, reloadPharos=True, fromDbPharos=True):
-        """Export FASTA target files (and load Pharos from source)"""
+        """Export FASTA files for each target resource (and load Pharos from source).
+
+            --> Creates files:
+                - CACHE/SAbDab-features/sabdab-data.json  (e.g., (1074) Thera-SAbDab data records and (31152) SAbDab assignments)
+                - CACHE/FASTA/sabdab-targets.fa  (e.g., SAbDab 2120 fasta sequences)
+                - CACHE/FASTA/card-targets.fa
+                - CACHE/FASTA/card-targets-taxon.tdd  (TDD is taxonomy database)
+                - CACHE/FASTA/chembl-targets.fa
+                - CACHE/FASTA/chembl-targets-taxon.tdd
+                - CACHE/FASTA/pharos-targets.fa
+                - CACHE/FASTA/pharos-targets-taxon.tdd
+                - ...and more...
+        """
         logger.info("Running exportFasta...")
         ok = False
         try:
@@ -155,7 +183,16 @@ class ProteinTargetSequenceExecutionWorkflow(object):
         return ok
 
     def createSearchDatabases(self):
-        """Create search databases"""
+        """Create search databases for each target resource.
+
+            --> Creates directories/files:
+                - CACHE/sequence-databases/sabdab/    (SABDAB is only non-taxonomy database; no TDD)
+                - CACHE/sequence-databases/card/      (incl. taxonomy database, and all below)
+                - CACHE/sequence-databases/chembl/    (file 'chembl' contains all sequences)
+                - CACHE/sequence-databases/pharos/
+                - CACHE/sequence-databases/pdbprent/
+                - ...and more...
+        """
         logger.info("Running createSearchDatabases...")
         ok = False
         try:
@@ -166,15 +203,50 @@ class ProteinTargetSequenceExecutionWorkflow(object):
         return ok
 
     def searchDatabases(self):
-        """Search sequence databases"""
+        """Perform a search for each target resource DB against the 'pdbprent' database.
+
+        Args (provided below):
+            maxSeqs (int): Maximum results per query sequence allowed to pass the prefilter (affects sensitivity).
+                           This corresponds to the '--max-seqs' flag in MMseqs2 (default: 300).
+                           The higher the value, the fewer sequences will be filtered out (in cases of high sequence redundancy).
+                           Note: Use with caution--this can dramatically increase disk usage (https://github.com/soedinglab/MMseqs2/wiki#disk-space).
+
+            --> Creates files:
+                - CACHE/sequence-search-results/card-vs-pdbprent-filtered-results.json  (number of keys represents number of query sequences with matches)
+                - CACHE/sequence-search-results/card-vs-pdbprent-raw-results.json
+                - CACHE/sequence-search-results/card-vs-pdbprent-raw-results.txt
+                - CACHE/sequence-search-results/chembl-vs-pdbprent-filtered-results.json
+                - CACHE/sequence-search-results/chembl-vs-pdbprent-raw-results.json
+                - CACHE/sequence-search-results/chembl-vs-pdbprent-raw-results.txt
+                - CACHE/sequence-search-results/pharos-vs-pdbprent-filtered-results.json
+                - CACHE/sequence-search-results/pharos-vs-pdbprent-raw-results.json
+                - CACHE/sequence-search-results/pharos-vs-pdbprent-raw-results.txt
+                - CACHE/sequence-search-results/sabdab-vs-pdbprent-filtered-results.json
+                - CACHE/sequence-search-results/sabdab-vs-pdbprent-raw-results.json
+                - CACHE/sequence-search-results/sabdab-vs-pdbprent-raw-results.txt
+                - ...and more...
+        """
         logger.info("Running searchDatabases...")
         ok = False
         try:
             ptsW = ProteinTargetSequenceWorkflow(self.__cfgOb, self.__cachePath)
             ok1 = ptsW.search(
-                referenceResourceName="pdbprent", resourceNameList=["sabdab", "card", "drugbank", "chembl", "pharos"], identityCutoff=0.95, sensitivity=4.5, timeOutSeconds=1000
+                referenceResourceName="pdbprent",
+                resourceNameList=["sabdab", "drugbank", "chembl", "pharos"],
+                identityCutoff=0.95,
+                sensitivity=4.5,
+                timeOutSeconds=1000,
+                maxSeqs=750,  # number of seqs permitted past the prefilter (default 300; use caution when increasing w.r.t. disk usage)
+            )
+            ok2 = ptsW.search(
+                referenceResourceName="pdbprent",
+                resourceNameList=["card"],
+                identityCutoff=0.95,
+                sensitivity=4.5,
+                timeOutSeconds=1000,
+                useBitScore=True,
+                maxSeqs=750,
             )
-            ok2 = ptsW.search(referenceResourceName="pdbprent", resourceNameList=["card"], identityCutoff=0.95, sensitivity=4.5, timeOutSeconds=1000, useBitScore=True)
             ok = ok1 and ok2
         except Exception as e:
             logger.exception("Failing with %s", str(e))
@@ -237,29 +309,47 @@ class ProteinTargetSequenceExecutionWorkflow(object):
 def fullWorkflow():
     """Entry point for the full targets sequence and cofactor update workflow."""
     ptsWf = ProteinTargetSequenceExecutionWorkflow()
+    #
+    # Fetch taxonomy data
     ok = ptsWf.cacheTaxonomy()
     logger.info("cacheTaxonomy status %r", ok)
     ok = ptsWf.updateUniProtTaxonomy() and ok
     logger.info("updateUniProtTaxonomy status %r", ok)
+    #
+    # Fetch all PDB entity sequences from 'pdbx_core_polymer_entity'
     ok = ptsWf.fetchProteinEntityData() and ok
     logger.info("fetchProteinEntityData status %r", ok)
+    #
+    # Fetch/prepare mapping of all chemical references from 'bird_chem_comp_core' (for DrugBank and ChEMBL)
     ok = ptsWf.fetchChemicalReferenceMappingData() and ok
     logger.info("fetchChemicalReferenceMappingData status %r", ok)
+    #
+    # Fetch/prepare mapping of all polymer_entities and their associated ligands (e.g., "2E1B_1": ["ZN"])
     ok = ptsWf.fetchLigandNeighborMappingData() and ok
     logger.info("fetchLigandNeighborMappingData status %r", ok)
+    #
+    # Export FASTA files for each target resource
     ok = ptsWf.exportFasta() and ok
     logger.info("exportFasta status %r", ok)
+    #
+    # Create search databases for each target resource
     ok = ptsWf.createSearchDatabases() and ok
     logger.info("createSearchDatabases status %r", ok)
+    #
+    # Search each target resource DB against the 'pdbprent' database...
     ok = ptsWf.searchDatabases() and ok
     logger.info("searchDatabases status %r", ok)
+    #
     ok = ptsWf.buildFeatures() and ok
     logger.info("buildFeatures status %r", ok)
+    #
     # To rebuild ChEMBL-target-activity data from scratch (non-incremental), set skip=None in fetchTargetActivityDataMulti()
     ok = ptsWf.buildActivityData() and ok
     logger.info("buildActivityData status %r", ok)
+    #
     ok = ptsWf.buildCofactorData() and ok
     logger.info("buildCofactorData status %r", ok)
+    #
     ptsWf.resourceCheck()
     return ok
 

{rcsb_workflow-0.45 → rcsb_workflow-0.47}/rcsb/workflow/targets/ProteinTargetSequenceWorkflow.py

@@ -17,6 +17,7 @@
 #   6-Jul-2023 aae Don't overwrite Buildlocker files if there is no data
 #  20-Aug-2024 dwp Add step for loading target cofactor data to MongoDB
 #  27-Aug-2024 dwp Update usage of CARDTargetOntologyProvider
+#  10-Dec-2024 dwp Add support for 'max-seqs' flag in mmseqs search
 ##
 __docformat__ = "google en"
 __author__ = "John Westbrook"
@@ -252,7 +253,7 @@ class ProteinTargetSequenceWorkflow(object):
             retOk = False
         return retOk
 
-    def search(self, referenceResourceName, resourceNameList=None, identityCutoff=0.90, timeOutSeconds=10, sensitivity=4.5, useBitScore=False, formatOutput=None):
+    def search(self, referenceResourceName, resourceNameList=None, identityCutoff=0.90, timeOutSeconds=10, sensitivity=4.5, useBitScore=False, formatOutput=None, maxSeqs=300):
         """Search for similar sequences in the reference resource and the input sequence resources.
 
         Args:
@@ -264,6 +265,7 @@ class ProteinTargetSequenceWorkflow(object):
             useBitScore (bool, optional): use bitscore value as an additional comparison threshold. Defaults to False.
             formatOutput(str, optional):  mmseq2 search fields exported. Defaults to "query,target,taxid,taxname,pident,alnlen,mismatch,
                                                                          gapopen,qstart,qend,tstart,tend,evalue,raw,bits,qlen,tlen,qaln,taln,cigar".
+            maxSeqs (int): Maximum results per query sequence allowed to pass the prefilter (affects sensitivity). Defaults to 300.
 
         Returns:
              bool: True for success or False otherwise
@@ -275,7 +277,14 @@ class ProteinTargetSequenceWorkflow(object):
                 continue
             startTime = time.time()
             ok = self.__searchSimilar(
-                referenceResourceName, resourceName, identityCutoff=identityCutoff, timeOut=timeOutSeconds, sensitivity=sensitivity, useBitScore=useBitScore, formatOutput=formatOutput
+                referenceResourceName,
+                resourceName,
+                identityCutoff=identityCutoff,
+                timeOut=timeOutSeconds,
+                sensitivity=sensitivity,
+                useBitScore=useBitScore,
+                formatOutput=formatOutput,
+                maxSeqs=maxSeqs,
             )
             logger.info(
                 "Completed searching %s targets (status %r) (cutoff=%r) at %s (%.4f seconds)",
@@ -289,7 +298,7 @@ class ProteinTargetSequenceWorkflow(object):
         #
         return retOk
 
-    def __searchSimilar(self, referenceResourceName, resourceName, identityCutoff=0.90, timeOut=10, sensitivity=4.5, useBitScore=False, formatOutput=None):
+    def __searchSimilar(self, referenceResourceName, resourceName, identityCutoff=0.90, timeOut=10, sensitivity=4.5, useBitScore=False, formatOutput=None, maxSeqs=300):
         """Search for similar sequences in reference resource and input resources"""
         try:
             resultDirPath = self.__getResultDirPath()
@@ -302,7 +311,7 @@ class ProteinTargetSequenceWorkflow(object):
             rawPath = self.__getSearchResultPath(resourceName, referenceResourceName)
             resultPath = self.__getFilteredSearchResultPath(resourceName, referenceResourceName)
             ok = mmS.searchDatabase(
-                resourceName, seqDbTopPath, referenceResourceName, rawPath, minSeqId=identityCutoff, timeOut=timeOut, sensitivity=sensitivity, formatOutput=formatOutput
+                resourceName, seqDbTopPath, referenceResourceName, rawPath, minSeqId=identityCutoff, timeOut=timeOut, sensitivity=sensitivity, formatOutput=formatOutput, maxSeqs=maxSeqs
             )
             #
             if taxonPath and mU.exists(taxonPath):

{rcsb_workflow-0.45 → rcsb_workflow-0.47/rcsb.workflow.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rcsb.workflow
-Version: 0.45
+Version: 0.47
 Summary: RCSB Python data processing and ETL/ELT workflow entry points
 Home-page: https://github.com/rcsb/py-rcsb_workflow
 Author: John Westbrook
@@ -12,15 +12,16 @@ Classifier: Natural Language :: English
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: OpenEye-toolkits>=2020.2.2
+Requires-Dist: OpenEye-toolkits>=2024.1.1
 Requires-Dist: rcsb.utils.chem>=0.79
 Requires-Dist: rcsb.utils.seq>=0.82
 Requires-Dist: rcsb.utils.targets>=0.82
-Requires-Dist: rcsb.utils.seqalign>=0.29
-Requires-Dist: rcsb.exdb>=1.23
+Requires-Dist: rcsb.utils.seqalign>=0.31
+Requires-Dist: rcsb.exdb>=1.25
 Requires-Dist: rcsb.utils.io>=1.48
 Provides-Extra: dev
 Requires-Dist: check-manifest; extra == "dev"

{rcsb_workflow-0.45 → rcsb_workflow-0.47}/rcsb.workflow.egg-info/requires.txt

@@ -1,9 +1,9 @@
-OpenEye-toolkits>=2020.2.2
+OpenEye-toolkits>=2024.1.1
 rcsb.utils.chem>=0.79
 rcsb.utils.seq>=0.82
 rcsb.utils.targets>=0.82
-rcsb.utils.seqalign>=0.29
-rcsb.exdb>=1.23
+rcsb.utils.seqalign>=0.31
+rcsb.exdb>=1.25
 rcsb.utils.io>=1.48
 
 [dev]

{rcsb_workflow-0.45 → rcsb_workflow-0.47}/requirements.txt

@@ -1,8 +1,8 @@
 --extra-index-url https://pypi.anaconda.org/OpenEye/simple
-OpenEye-toolkits >= 2020.2.2
+OpenEye-toolkits >= 2024.1.1
 rcsb.utils.chem >= 0.79
 rcsb.utils.seq >= 0.82
 rcsb.utils.targets >= 0.82
-rcsb.utils.seqalign >= 0.29
-rcsb.exdb >= 1.23
+rcsb.utils.seqalign >= 0.31
+rcsb.exdb >= 1.25
 rcsb.utils.io >= 1.48

{rcsb_workflow-0.45 → rcsb_workflow-0.47}/setup.py

@@ -45,7 +45,8 @@ setup(
         "License :: OSI Approved :: Apache Software License",
         "Programming Language :: Python",
         "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.8",
+        "Programming Language :: Python :: 3.9",
+        "Programming Language :: Python :: 3.10",
     ],
     # entry_points={"console_scripts": ["cactvs_annotate_mol=rcsb.workflow.cactvsAnnotateMol:main"]},
     #  The following is somewhat flakey --