Multiple configs

README.md

@@ -224,59 +224,6 @@ the data that you want to integrate.
 ./weave.py –oncokb /path_to_file/test_genomics_oncokbannotation.csv
 ```
 
-
-### Gene Ontology adapter
-
-**Gene Ontology** is one of the biggest biomedical databases. The described
-adapter helps to integrate the data about the molecular function of the gene
-product, as well as the biological process in which these genes are involved.
-
-- Molecular function: GO annotations that have relation type `enabled`
-  or `contributes_to`.
-- Biological process: GO annotations that have relation type `involved_in`.
-
-**To integrate the data, three files are necessary:**
-- `--gene_ontology` option for GO annotations in GAF format  [Download GO annotations](http://current.geneontology.org/products/pages/downloads.html)
-- `--gene_ontology_owl` option for GO ontology in OWL format [Download GO ontology](https://geneontology.org/docs/download-ontology/)
-- `--gene_ontology_genes` option for the list of genes for which we want to
-  integrate the GO annotations (example in adapters/Hugo_Symbol_genes.conf file,
-  by default = list of genes from OncoKB database).
-
-**Example of use:**
-
-``` sh
-./weave.py --gene_ontology /path_to_file/goa_human.gaf --gene_ontology_owl /path_to_file/go.owl --gene_ontology_genes /path_to_file/Hugo_Symbol_genes.conf
-```
-
-If you want to integrate annotations with another type of relations, you can
-modify the `adapters/gene_ontology.py` file by adding the next code in the
-**class Gene_ontology** (example for the `involved_in` edge type):
-
-``` python
-# Create new columns that depends on edge type.
-df['GO_involved_in'] = None
-
-# Cut df to include only edge type that we have chosen and annotations
-# for genes from OncoKB.
-df = df[((df['Qualifier'].isin(['enables', 'involved_in', 'contributes_to'])) &
-         (df['DB_Object_Symbol'].isin(included_genes)))]
-```
-Also, you need to add code in `separate_edges_types` method:
-
-``` sh
-# Function to copy GO_term to related column for future ontoweaver mapping
-# based on Qualifier column (relation type).
-   def separate_edges_types(row):
-        if row['Qualifier'] == 'enables':
-            row['GO_enables'] = row['GO_term']
-        elif row['Qualifier'] == 'involved_in':
-            row['GO_involved_in'] = row['GO_term']
-```
-
-Finally, you need to specify the node and edge types in the `gene_ontology.yaml`
-for `GO_involved_in` column.
-
-
 ### Open Targets adapter
 
 Open Targets is a public database that aims to systematically identify and

config/biopathnet.yaml

@@ -9,9 +9,11 @@ biocypher:
         root_node: entity
 
 biopathnet:
-  file_format: txt
+  file_format: txt:bn 
   entity_types_file_stem: entity_types
   entity_names_file_stem: entity_names
   background_graph_file_stem: brg
   skg_file_stem: skg
+  targeted_relation: "(alteration, variant biomarker for treatment, drug)"
+  include_properties: False
 

config/owl.yaml

@@ -0,0 +1,16 @@
+biocypher:
+    debug: false
+    offline: true
+    dbms: owl
+
+    # Ontology configuration
+    head_ontology:
+        url: https://github.com/biolink/biolink-model/raw/v3.2.1/biolink-model.owl.ttl
+        root_node: entity
+
+owl:
+    edge_model: ObjectProperty
+    file_format: turtle
+    labels_order: "Ascending" # Default: From more specific to more generic.
+    node_labels_order: "Ascending" # Default: use labels_order.
+    edge_labels_order: "Leaves"

config/schema.yaml

@@ -4,12 +4,11 @@
 
 # Defined in alphabetical order
 
-alteration:
+short mutation:
     is_a: sequence variant
     represented_as: node
-    label_in_input: alteration
+    label_in_input: short_mutation
     properties:
-        gene_symbol_alteration: str
         citation_PM_ids: str
         consequence: str
         homogenous: str
@@ -23,7 +22,47 @@ alteration:
         refCount: int64
         altCount: int64
         expressed: bool
-        ensembl_id_alteration: str
+        # ensembl_id_alteration: str
+
+copy number amplification:
+    is_a: sequence variant
+    represented_as: node
+    label_in_input: copy_number_amplification
+    properties:
+        citation_PM_ids: str
+        consequence: str
+        homogenous: str
+        mutation_effect_description: str
+        data_source: str
+        oncogenic: str
+        reference_genome: str
+        tumor_type: str
+        tumor_type_summary: str
+        variant_summary: str
+        refCount: int64
+        altCount: int64
+        expressed: bool
+        # ensembl_id_alteration: str
+
+structural variant:
+    is_a: sequence variant
+    represented_as: node
+    label_in_input: structural_variant
+    properties:
+        citation_PM_ids: str
+        consequence: str
+        homogenous: str
+        mutation_effect_description: str
+        data_source: str
+        oncogenic: str
+        reference_genome: str
+        tumor_type: str
+        tumor_type_summary: str
+        variant_summary: str
+        refCount: int64
+        altCount: int64
+        expressed: bool
+        # ensembl_id_alteration: str
 
 disease:
     represented_as: node
@@ -202,6 +241,12 @@ protein:
         ncbi_tax_id: str
         data_source: str
 
+# Treatment
+
+treatment:
+    represented_as: node
+    input_label: treatment
+
 ########################
 # EDGES
 ########################
@@ -212,7 +257,7 @@ protein:
 
 ### CARRIES
 
-# To allow queries for patient carrying samples, and samples carrying alterations,
+# To allow queries for patient carrying samples, and samples carrying variants,
 # without mixing with "effects" causes.
 carries:
     is_a: causes
@@ -232,12 +277,12 @@ patient carries sample:
         data_source: str
         edglelabel: str
 
-sample carries alteration:
+sample carries variant:
     is_a: carries
     represented_as: edge
-    label_in_input: sample_carries_alteration
+    label_in_input: sample_carries_variant
     source: sample
-    target: alteration
+    target: sequence variant
     properties:
         data_source: str
         edglelabel: str
@@ -246,16 +291,16 @@ sample carries alteration:
 
 # A gene is linked to its gene status (gain or loss of function),
 # which are represented as nodes, so as to allow a causal path
-# to go through alteration -> gene status -> transcript activity.
+# to go through variant -> gene status -> transcript activity.
 # Hence, outcomes have at least two instances:
 # - Gene:GoF, and
 # - Gene:LoF.
 
-alteration causes gene status:
+variant causes gene status:
     is_a: causes
     represented_as: edge
-    label_in_input: alteration_causes_gene_status
-    source: alteration
+    label_in_input: variant_causes_gene_status
+    source: sequence variant
     target: gene status
     properties:
         data_source: str
@@ -267,13 +312,17 @@ alteration causes gene status:
 # as predictive markers for treatment response, 
 # based on clinical evidence categorized by evidence levels. 
 
-alteration biomarker for drug:
-    is_a: biomarker for
+variant biomarker for treatment:
+    is_a: sequence variant modulates treatment association
     represented_as: edge
-    label_in_input: alteration_biomarker_for_drug
-    source: alteration
-    target: drug
-    properties:
+    label_in_input: variant_biomarker_for_treatment
+    source: sequence variant
+    target: treatment
+    properties:
+        level_of_evidence: str
+        cgi_level: str
+        citations: str
+        tumorType: str
         data_source: str
         edglelabel: str
 
@@ -302,10 +351,6 @@ gene status affects gene:
 
 
 ### GO -- TO BE FIXED
-# annotation: 
-#     is_a: named thing
-#     represented_as: node
-#     label_in_input: annotation
 
 biological process:
     is_a: named thing
@@ -314,20 +359,6 @@ biological process:
     properties:
         data_source: str
 
-# annotation for gene: 
-#     is_a: association
-#     represented_as: edge
-#     label_in_input: annotation_for_gene
-#     source: annotation
-#     target: gene
-
-# involved in: 
-#     is_a: association
-#     represented_as: edge
-#     label_in_input: involved_in
-#     source: annotation
-#     target: biological process
-
 gene to biological process:
     is_a: association
     represented_as: edge
@@ -345,9 +376,10 @@ biological process to gene:
     source: biological process
     target: gene
     properties:
-        # edglelabel: str
         data_source: str
 
+### FUNCTIONAL PROTEIN PROTEIN INTERACTIONS
+
 undirected molecular interaction:
     is_a: pairwise molecular interaction
     represented_as: edge
@@ -447,16 +479,28 @@ inhibition:
         extra_attrs: str
         evidences: str
 
+### TRASNCRIPT TO GENE RELATIONSHIP
+
 transcript to gene relationship:
     # is_a: transcript to gene relationship
     represented_as: edge
     input_label: transcript_to_gene_relationship
     properties:
         data_source: str
 
+### DRUG HAS TARGET
+
 drug has target:
     is_a: drug to gene association
     represented_as: edge
     label_in_input: drug_has_target
     properties:
         data_source: str
+
+treatment has part drug:
+    is_a: association
+    represented_as: edge
+    label_in_input: treatment_has_part_drug
+    properties:
+        data_source: str
+

make.sh

@@ -65,7 +65,7 @@ echo "Activate virtual environment..." >&2
 source $(dirname $(uv python find))/activate
 
 
-if [[ "$2" == "config/neo4j.yaml" ]] ; then
+if [[ "$CONFIG" == "config/neo4j.yaml" ]] ; then
     echo "Stop Neo4j server..." >&2
     neo_version=$(neo4j-admin --version | cut -d. -f 1)
     if [[ "$neo_version" -eq 4 ]]; then
@@ -78,23 +78,23 @@ fi
 
 echo "Weave data..." >&2
 
+echo "CONFIG = $CONFIG" >&2
+
 cmd="uv run python3 ${py_args} $script_dir/weave.py \
-    --config $CONFIG \
+    --copy-number-amplifications-external   $decider_dir/cnas_external.csv  \
     --short-mutations-local                 $decider_dir/short_mutations_local.csv  \
     --short-mutations-external              $decider_dir/short_mutations_external.csv  \
     --copy-number-amplifications-local      $decider_dir/cnas_local.csv \
-    --copy-number-amplifications-external   $decider_dir/cnas_external.csv  \
-    --omnipath-networks                     $data_dir/omnipath_networks/omnipath_webservice_interactions__latest.tsv.gz \
     --open-targets-drug-molecule            $data_dir/OT/drug_molecule/
     --open-targets-drug_mechanism_of_action $data_dir/OT/drug_mechanism_of_action/
     --open-targets-target                   $data_dir/OT/target/
+    --cgi                                   $decider_dir/treatments_cgi.csv \
+    --config $CONFIG \
     ${weave_args}" # \
+    # --omnipath-networks                     $data_dir/omnipath_networks/omnipath_networks_different_type_entity_type_source_and_entity_type_target_shorter.tsv \
+    # --structural-variants                   $decider_dir/structural_variants.xlsx  \
     # --clinical                              $data_dir/DECIDER/clinical/clinical_export.xlsx \
-    # --gene_ontology_genes        $data_dir/DECIDER/$data_version/OncoKB_gene_symbols.conf \
     # --oncokb                     $data_dir/DECIDER/$data_version/treatments.csv \
-    # --gene_ontology              $data_dir/GO/goa_human.gaf.gz \
-    # --gene_ontology_owl          $data_dir/GO/go.owl \
-    # --gene_ontology_reverse
 
 
 echo "Weaving command:" >&2