Add new study groups for human data split from Illumina runs

In Illumina sequencing runs of non-human DNA, assign any contamination identified as being human DNA to a special iRODS group named after the original study i.e. `ss_1000` -> `ss_1000_human`

src/npg_irods/illumina.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright © 2023 Genome Research Ltd. All rights reserved.
+# Copyright © 2023, 2024 Genome Research Ltd. All rights reserved.
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -234,7 +234,7 @@ def empty_acl(*args):
         for fc in flowcells:
             secondary_metadata.extend(sample_fn(fc.sample))
             secondary_metadata.extend(study_fn(fc.study))
-            acl.extend(acl_fn(fc.sample, fc.study, zone=zone))
+            acl.extend(acl_fn(fc.sample, fc.study, subset=c.subset, zone=zone))
 
     # Remove duplicates
     secondary_metadata = sorted(set(secondary_metadata))
@@ -248,7 +248,7 @@ def empty_acl(*args):
         cons_update = ensure_consent_withdrawn(item)
     elif any(c.contains_nonconsented_human() for c in components):  # Illumina specific
         log.info("Non-consented human data", path=item)
-        xahu_update = ensure_consent_withdrawn(item)
+        xahu_update = update_permissions(item, acl)
     else:
         perm_update = update_permissions(item, acl)
 

src/npg_irods/metadata/lims.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright © 2021, 2022, 2023 Genome Research Ltd. All rights reserved.
+# Copyright © 2021, 2022, 2023, 2024 Genome Research Ltd. All rights reserved.
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -38,6 +38,7 @@
 from npg_irods.db.mlwh import Sample, Study
 from npg_irods.metadata.common import (
     SeqConcept,
+    SeqSubset,
     ensure_avus_present,
     avu_if_value,
 )
@@ -168,7 +169,9 @@ def make_reduced_study_metadata(study: Study) -> list[AVU]:
     return [avu_if_value(TrackedStudy.ID, study.id_study_lims)]
 
 
-def make_sample_acl(sample: Sample, study: Study, zone=None) -> list[AC]:
+def make_sample_acl(
+    sample: Sample, study: Study, subset: SeqSubset = None, zone=None
+) -> list[AC]:
     """Returns an ACL for a given Sample in a Study.
 
     This method takes into account all factors influencing access control, which are:
@@ -187,12 +190,19 @@ def make_sample_acl(sample: Sample, study: Study, zone=None) -> list[AC]:
         sample: A sample, which will be used to confirm consent, which modifies the
                 ACL.
         study: A study, which will provide permissions for the ACL.
+        subset: Subset of sequence reads.
         zone: The iRODS zone.
 
     Returns:
         An ACL
     """
-    irods_group = f"{STUDY_IDENTIFIER_PREFIX}{study.id_study_lims}"
+    if subset is not None and subset is subset.XAHUMAN:
+        return []
+
+    if subset is not None and subset is subset.HUMAN:
+        irods_group = f"{STUDY_IDENTIFIER_PREFIX}{study.id_study_lims}_human"
+    else:
+        irods_group = f"{STUDY_IDENTIFIER_PREFIX}{study.id_study_lims}"
     perm = Permission.NULL if sample.consent_withdrawn else Permission.READ
 
     return [AC(irods_group, perm, zone=zone)]

tests/conftest.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright © 2020, 2022, 2023 Genome Research Ltd. All rights reserved.
+# Copyright © 2020, 2022, 2023, 2024 Genome Research Ltd. All rights reserved.
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -1029,6 +1029,17 @@ def illumina_synthetic_irods(tmp_path):
             *run_pos,
             AVU(tag, 1),
         ),
+        "12345/12345#1_xahuman.cram": (
+            AVU(idp, "bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb"),
+            AVU(
+                cmp, '{"id_run":12345, "position":1, "tag_index":1, "subset":"xahuman"}'
+            ),
+            AVU(
+                cmp, '{"id_run":12345, "position":2, "tag_index":1, "subset":"xahuman"}'
+            ),
+            *run_pos,
+            AVU(tag, 1),
+        ),
         "12345/12345#2.cram": (
             AVU(idp, "0b3bd00f1d186247f381aa87e213940b8c7ab7e5"),
             AVU(cmp, '{"id_run":12345, "position":1, "tag_index":2}'),

tests/test_illumina.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright © 2023 Genome Research Ltd. All rights reserved.
+# Copyright © 2023, 2024 Genome Research Ltd. All rights reserved.
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -412,7 +412,7 @@ def test_updates_changed_study_permissions(
 
     @m.context("When data are multiplexed")
     @m.context("When data contain a human subset")
-    @m.it("Removes managed access permissions")
+    @m.it("Updates managed access permissions to restricted human access group")
     def test_updates_human_permissions_mx(
         self, illumina_synthetic_irods, illumina_synthetic_mlwh
     ):
@@ -425,7 +425,10 @@ def test_updates_human_permissions_mx(
             AC("irods", perm=Permission.OWN, zone=zone),
             AC("ss_4000", perm=Permission.READ, zone=zone),
         ]
-        new_permissions = [AC("irods", perm=Permission.OWN, zone=zone)]
+        new_permissions = [
+            AC("irods", perm=Permission.OWN, zone=zone),
+            AC("ss_4000_human", perm=Permission.READ, zone=zone),
+        ]
 
         for obj in [DataObject(path), DataObject(qc_path)]:
             obj.add_permissions(*old_permissions)

tests/test_locate_data_objects.py

@@ -20,6 +20,7 @@ def test_illumina_updates(
                 "12345#1.cram",
                 "12345#1_human.cram",
                 "12345#1_phix.cram",
+                "12345#1_xahuman.cram",
                 "12345#2.cram",
                 "12345#888.cram",
                 "12345.cram",

tests/test_pacbio.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright © 2023 Genome Research Ltd. All rights reserved.
+# Copyright © 2023, 2024 Genome Research Ltd. All rights reserved.
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by