1.6.5

first multi-variant match etc.; see http://bycon.progenetix.org/changes/

bycon/__init__.py

@@ -17,6 +17,7 @@
     from beacon_response_generation import *
     from bycon_helpers import *
     from cgi_parsing import *
+    from cytoband_parsing import *
     from dataset_parsing import *
     from genome_utils import *
     from handover_generation import *

bycon/definitions/argument_definitions.yaml

@@ -78,15 +78,17 @@ include_resultset_responses:
 
 dataset_ids:
   type: array
-  items: string
+  items:
+    type: string
   cmdFlags:
     - -d
     - --datasetIds
   description: dataset ids
 
 filters:
   type: array
-  items: string
+  items:
+    type: string
   cmdFlags:
     - --filters
   description: prefixed filter values, comma concatenated
@@ -202,9 +204,11 @@ variant_max_length:
   description: 'variantMaxLength: The maximum variant length in bases for e.g. CNV queries.'
 
 gene_id:
-  type: string
+  type: array
+  items:
+    type: string
+    pattern: '^\w+?(\w+?(\-\w+?)?)?$'
   db_key: Null
-  pattern: '^\w+?(\w+?(\-\w+?)?)?$'
   cmdFlags:
     - --geneId
   description: gene id
@@ -235,6 +239,18 @@ genomic_allele_short_form:
 # variant parameters - non-standard
 # ------------------------------------------------------------------------------
 
+variant_query_digests:
+  type: array
+  db_key: Null
+  items:
+    type: string
+    pattern: '^(?:chro?)?([12]?[\dXY]):(\d+?(?:-\d+?)?)(?:--(\d+?(?:-\d+?)?))?(?::([\w\:\>]+?))?$'
+  cmdFlags:
+    - --variantQueryDigests
+  examples:
+    - "9:9000001-21975098--21967753-24000000:DEL"
+  description: 'Variant query digest-style short form'
+
 variant_internal_id:
   type: string
   db_key: variant_internal_id 
@@ -281,38 +297,48 @@ id:
 
 ids:
   type: array
-  items: string
+  items:
+    type: string
+    pattern: '^\w[\w:-]+\w$'
   cmdFlags:
     - --ids
   description: 'One or more ids; this parameter only makes sense for specific REST entry types'
 
 biosample_ids:
   type: array
-  items: string
+  items:
+    type: string
+    pattern: '^\w[\w:-]+\w$'
   byc_entity: biosample
   cmdFlags:
     - --biosampleIds
   description: biosample ids
 
 analysis_ids:
   type: array
-  items: string
+  items:
+    type: string
+    pattern: '^\w[\w:-]+\w$'
   byc_entity: analysis
   cmdFlags:
     - --analysisIds
   description: callset / analysis ids
 
 individual_ids:
   type: array
-  items: string
+  items:
+    type: string
+    pattern: '^\w[\w:-]+\w$'
   byc_entity: individual
   cmdFlags:
     - --individualIds
   description: subject ids
 
 variant_ids:
   type: array
-  items: string
+  items:
+    type: string
+    pattern: '^\w[\w:-]+\w$'
   byc_entity: genomicVariant
   cmdFlags:
     - --variantIds
@@ -465,14 +491,16 @@ source:
 
 delivery_keys:
   type: array
-  items: string
+  items:
+    type: string
   cmdFlags:
     - --deliveryKeys
   description: delivery keys
 
 collation_types:
   type: array
-  items: string
+  items:
+    type: string
   cmdFlags:
     - --collationTypes
   description: selected collation types, e.g. "EFO"
@@ -485,7 +513,8 @@ with_samples:
 
 selected_beacons:
   type: array
-  items: string
+  items:
+    type: string
 
 # ------------------------------------------------------------------------------
 # genomic parameters
@@ -499,17 +528,21 @@ genome_binning:
   description: "one of the predefined genome binning keys - default 1Mb"
 
 cyto_bands:
-  type: string
+  type: array
   db_key: Null
-  pattern: ^(?:chro?)?([12]?[\dXY])([pq](?:(?:ter)|(?:cen)|(?:[1-4](?:\d(?:\.\d\d*?)?)?)?))?\-?([pq](?:(?:cen)|(?:ter)|(?:[1-4](?:\d(?:\.\d\d*?)?)?)?))?$
+  items:
+    type: string
+    pattern: ^(?:chro?)?([12]?[\dXY])([pq](?:(?:ter)|(?:cen)|(?:[1-4](?:\d(?:\.\d\d*?)?)?)?))?\-?([pq](?:(?:cen)|(?:ter)|(?:[1-4](?:\d(?:\.\d\d*?)?)?)?))?$
   cmdFlags:
     - --cytoBands
   description: cytobands, e.g. 8q21q24.1
 
 chro_bases:
-  type: string
+  type: array
+  items:
+    type: string
+    pattern: ^(chro?)?([12]?[\dXY])\:(\d+?)(\-(\d+?))?$
   db_key: Null
-  pattern: ^(chro?)?([12]?[\dXY])\:(\d+?)(\-(\d+?))?$
   cmdFlags:
     - --chroBases
   description: only for the cytoband converter ... e.g. 8:0-120000000

bycon/definitions/datatable_mappings.yaml

@@ -110,6 +110,17 @@ definitions:
         db_key: references.cbioportal.label
         beacon_key: False
         compact: True
+      tcgaproject_id:
+        type: string
+        db_key: references.tcgaproject.id
+        indexed: True
+        beacon_key: False
+        compact: True
+      tcgaproject_label:
+        type: string
+        db_key: references.tcgaproject.label
+        beacon_key: False
+        compact: True
       external_references:
         type: array
         beacon_key: biosamples.externalReferences

bycon/definitions/variant_request_definitions.yaml

@@ -18,12 +18,26 @@ request_pars:
   - gene_id
   - aminoacid_change
   - genomic_allele_short_form
+  - cyto_bands
+  - variant_query_digests
 
 # As long as there are no variant query schemas to define valid Beacon
 # variant parameter combinations the `request_types` here provide sets
 # of combinations valid at least for `bycon` Progenetix etc. data.
 request_types:
 
+  variantQueryDigestsRequest:
+    all_of:
+      - variant_query_digests
+
+  cytoBandRequest:
+    all_of:
+      - cyto_bands
+    optional:
+      - variant_type
+      - variant_min_length
+      - variant_max_length
+
   aminoacidChangeRequest:
     all_of:
       - aminoacid_change
@@ -34,14 +48,14 @@ request_types:
     all_of:
       - genomic_allele_short_form
 
-  variantTypeRequest:
-    description: >-
-      This dangerous query only uses a variant type, e.g. to limit an "all variants"
-      retrieval to SNVs or CNVs.
-      TODO: Add some private parameter to make this work on ly for special
-      circumstances... May be used w/ a combination of filter length >0 check.
-    all_of:
-      - variant_type
+  # variantTypeRequest:
+  #   description: >-
+  #     This dangerous query only uses a variant type, e.g. to limit an "all variants"
+  #     retrieval to SNVs or CNVs.
+  #     TODO: Add some private parameter to make this work only for special
+  #     circumstances... May be used w/ a combination of filter length > 0 check.
+  #   all_of:
+  #     - variant_type
 
   variantBracketRequest:
     description: >-
@@ -60,8 +74,8 @@ request_types:
       - start
       - end
       - reference_name
-      - variant_type
     optional:
+      - variant_type
       - variant_min_length
       - variant_max_length
 

bycon/definitions/variant_type_definitions.yaml

@@ -65,6 +65,27 @@ EFO:0030066:
     - EFO:0030072
     - EFO:0030073      
 
+SO:0001743:
+  description: >-
+    This is an alternative to `EFO:0030067`. The SO definitions for CNVs are
+    slightly misleading (type of `sequence_length_alteration`) and do not
+    include the more specific child terms from EFO or VRS. 
+  variant_state:
+    id: SO:0001743
+    label: copy_number_loss
+  DUPDEL: DEL
+  variant_type_id: SO:0001743
+  variant_type: DEL
+  VRS_type: CopyNumberChange
+  VCF_symbolic_allele: <DEL>
+  cnv_dummy_value: -1
+  child_terms:
+    - EFO:0030067
+    - SO:0001743
+    - EFO:0030068
+    - EFO:0020073
+    - EFO:0030069
+
 EFO:0030067:
   variant_state:
     id: EFO:0030067
@@ -77,6 +98,7 @@ EFO:0030067:
   cnv_dummy_value: -1
   child_terms:
     - EFO:0030067
+    - SO:0001743
     - EFO:0030068
     - EFO:0020073
     - EFO:0030069
@@ -133,6 +155,28 @@ EFO:0030070:
   cnv_dummy_value: 0.58
   child_terms:
     - EFO:0030070
+    - SO:0001742
+    - EFO:0030071
+    - EFO:0030072
+    - EFO:0030073
+
+SO:0001742:
+  description: >-
+    This is an alternative to `EFO:0030070`. The SO definitions for CNVs are
+    slightly misleading (type of `sequence_length_alteration`) and do not
+    include the more specific child terms from EFO or VRS. 
+  variant_state:
+    id: SO:0001742
+    label: copy_number_gain
+  DUPDEL: DUP
+  variant_type_id: SO:0001742
+  variant_type: DUP
+  VRS_type: CopyNumberChange
+  VCF_symbolic_allele: <DUP>
+  cnv_dummy_value: 0.58
+  child_terms:
+    - EFO:0030070
+    - SO:0001742
     - EFO:0030071
     - EFO:0030072
     - EFO:0030073
@@ -192,6 +236,7 @@ SO:0001059:
     - SO:0001483
     - SO:0002007
     - SO:0000667
+    - SO:0001743
 
 SO:0001483:
   variant_state:
@@ -231,3 +276,16 @@ SO:0000667:
   cnv_dummy_value: Null
   child_terms:
     - SO:0000667
+
+SO:0001413:
+  variant_state:
+    id: SO:0001413
+    label: translocation breakpoint
+  DUPDEL: Null
+  variant_type_id: SO:0001413
+  variant_type: BND
+  VRS_type: Null
+  VCF_symbolic_allele: <BND>
+  cnv_dummy_value: Null
+  child_terms:
+    - SO:0001413

bycon/lib/beacon_response_generation.py

@@ -431,7 +431,9 @@ def __return_filtering_terms(self):
 
         # TODO: This should be derived from some entity definitions
         # TODO: whole query generation in separate function ...
-        scopes = ["biosamples", "individuals", "analyses", "genomicVariations"]
+        # now added globally to filters since Q aggregation and 
+        # https://github.com/ga4gh-beacon/beacon-v2/pull/118
+        scopes = BYC.get("data_pipeline_entities", [])
         query = {}
         q_list = []
 
@@ -470,6 +472,8 @@ def __return_filtering_terms(self):
                 if ft_k and ft_s:
                     f_t.update({"target": f'{ft_s}.{ft_k}'})
 
+                f_t.update({"scopes": scopes})
+
                 lab = f_t.get("label")
                 if lab is None:
                     f_t.pop("label", None)
@@ -753,7 +757,9 @@ def __retrieve_variants_data(self):
             r_s_res = []
 
             if "variants._id" in ds_results:
-                for v_id in ds_results["variants._id"]["target_values"]:
+                q_v_s = ds_results["variants._id"]["target_values"]
+                q_v_s = return_paginated_list(q_v_s, self.skip, self.limit)
+                for v_id in q_v_s:
                     v = v_coll.find_one({"_id":v_id})
                     r_s_res.append(v)
                 self.datasets_data.update({ds_id: r_s_res})

bycon/lib/bycon_helpers.py

@@ -63,7 +63,8 @@ def refactor_value_from_defined_type(parameter, values, definition):
             return str(value)
     else:
         BYC["WARNINGS"].append(f"!!! Multiple values for {parameter} in request")
-        return '::'.join(values)
+        # re-joining ...
+        return ','.join(values)
 
 
 ################################################################################