Incorporate uniprot id into gene info (#160)

* Added uniprotkb_accession column to gene_info dataset * Updated gx validation for gene_info * Added the uniprot mapping file to the gene_info provenance * Updating gx validation * Updated gx validation to include uniprotkb_accessions * Finally got the gx validation to use the plural uniprotkb_accessions * Pytests passing * Removed rename_unknown_column() and related code * gene_info test_add_uniprot_id_to_gene_info_should_pass passing * uniprot fail test is failing * Removing testing yaml file that we do not need * Removing unnecessary changes to utils.py * Updated version number of syn54113663 to syn54113663.3 in config and test_config * Made the ensembl to uniprot test mapping file smaller * Improved uniprot code integration with expected flow of existing codebase * Removed unused regex for uniprot gx validation * Adding duplicate ensembl ID to uniprot testing file. Removing unused testing files. * Updating testing_gene_info docstring * Removed unused testing file * Removed unused variables in gene_info testing script --------- Co-authored-by: Beatriz Saldana <[email protected]>
Sage-Bionetworks · Nov 27, 2024 · ae4cf82 · ae4cf82
1 parent 6f41530
commit ae4cf82
Show file tree

Hide file tree

Showing 9 changed files with 198 additions and 49 deletions.
diff --git a/config.yaml b/config.yaml
@@ -169,6 +169,9 @@ datasets:
         - name: tep_adi_info
           id: syn51942280.3
           format: csv
+        - name: ensg_to_uniprot_mapping
+          id: syn54113663.3
+          format: tsv
       final_format: json
       custom_transformations:
         adjusted_p_value_threshold: 0.05
@@ -186,6 +189,8 @@ datasets:
         maximumlogcpm: max
         possible_replacement: ensembl_possible_replacements
         permalink: ensembl_permalink
+        uniprotkb_accession: uniprotkb_accessions
+        resource_identifier: ensembl_gene_id
       provenance:
         - syn25953363.13
         - syn12514826.5
@@ -199,6 +204,7 @@ datasets:
         - syn13363443.11
         - *genes_biodomains_provenance
         - syn51942280.3
+        - syn54113663.3
       agora_rename:
         symbol: hgnc_symbol
       destination: *dest

diff --git a/gx_suite_definitions/gene_info.ipynb b/gx_suite_definitions/gene_info.ipynb
@@ -343,6 +343,17 @@
     "validator.expect_column_values_to_match_json_schema(\"ensembl_info\", ensembl_info_schema)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# uniprotkb_accessions\n",
+    "validator.expect_column_values_to_be_of_type(\"uniprotkb_accessions\", \"list\")\n",
+    "validator.expect_column_values_to_have_list_members_of_type(column=\"uniprotkb_accessions\", member_type=\"str\")"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},

diff --git a/src/agoradatatools/etl/transform/gene_info.py b/src/agoradatatools/etl/transform/gene_info.py
@@ -24,6 +24,7 @@ def transform_gene_info(
     druggability = datasets["druggability"]
     biodomains = datasets["genes_biodomains"]
     tep_info = datasets["tep_adi_info"]
+    uniprot = datasets["ensg_to_uniprot_mapping"]
 
     # Modify the data before merging
 
@@ -129,6 +130,13 @@ def transform_gene_info(
         axis=1,
     )
 
+    # Collapse uniprot IDs into a list for each ensembl_gene_id
+    collapsed_uniprot = (
+        uniprot.groupby("ensembl_gene_id")["uniprotkb_accessions"]
+        .apply(list)
+        .reset_index()
+    )
+
     # Merge all the datasets
     gene_info = gene_metadata
 
@@ -142,6 +150,7 @@ def transform_gene_info(
         druggability,
         biodomains,
         tep_info,
+        collapsed_uniprot,
     ]:
         gene_info = pd.merge(
             left=gene_info,
@@ -245,6 +254,7 @@ def transform_gene_info(
             "is_tep",
             "resource_url",
             "ensembl_info",
+            "uniprotkb_accessions",
         ]
     ]
 

diff --git a/src/agoradatatools/great_expectations/gx/expectations/gene_info.json b/src/agoradatatools/great_expectations/gx/expectations/gene_info.json
@@ -514,25 +514,25 @@
       "kwargs": {
         "column": "biodomains",
         "list_members": [
-          "Myelination",
+          "Apoptosis",
+          "Vasculature",
+          "Lipid Metabolism",
+          "Oxidative Stress",
           "Mitochondrial Metabolism",
+          "APP Metabolism",
           "Proteostasis",
-          "Oxidative Stress",
           "DNA Repair",
-          "Metal Binding and Homeostasis",
-          "Structural Stabilization",
-          "Endolysosome",
+          "Synapse",
           "Immune Response",
-          "Autophagy",
           "Tau Homeostasis",
           "Cell Cycle",
-          "Vasculature",
-          "APP Metabolism",
-          "Lipid Metabolism",
-          "Apoptosis",
-          "Synapse",
           "Epigenetic",
-          "RNA Spliceosome"
+          "Metal Binding and Homeostasis",
+          "Endolysosome",
+          "Structural Stabilization",
+          "Myelination",
+          "RNA Spliceosome",
+          "Autophagy"
         ]
       },
       "meta": {}
@@ -635,6 +635,22 @@
         }
       },
       "meta": {}
+    },
+    {
+      "expectation_type": "expect_column_values_to_be_of_type",
+      "kwargs": {
+        "column": "uniprotkb_accessions",
+        "type_": "list"
+      },
+      "meta": {}
+    },
+    {
+      "expectation_type": "expect_column_values_to_have_list_members_of_type",
+      "kwargs": {
+        "column": "uniprotkb_accessions",
+        "member_type": "str"
+      },
+      "meta": {}
     }
   ],
   "ge_cloud_id": null,

diff --git a/test_config.yaml b/test_config.yaml
@@ -169,6 +169,9 @@ datasets:
         - name: tep_adi_info
           id: syn51942280.3
           format: csv
+        - name: ensg_to_uniprot_mapping
+          id: syn54113663.3
+          format: tsv
       final_format: json
       custom_transformations:
         adjusted_p_value_threshold: 0.05
@@ -186,6 +189,8 @@ datasets:
         maximumlogcpm: max
         possible_replacement: ensembl_possible_replacements
         permalink: ensembl_permalink
+        uniprotkb_accession: uniprotkb_accessions
+        resource_identifier: ensembl_gene_id
       provenance:
         - syn25953363.13
         - syn12514826.5
@@ -199,6 +204,7 @@ datasets:
         - syn13363443.11
         - *genes_biodomains_provenance
         - syn51942280.3
+        - syn54113663.3
       agora_rename:
         symbol: hgnc_symbol
       destination: *dest

diff --git a/tests/test_assets/gene_info/input/ensg_to_uniprot_mapping_good.tsv b/tests/test_assets/gene_info/input/ensg_to_uniprot_mapping_good.tsv
@@ -0,0 +1,20 @@
+uniprotkb_accessions	ensembl_gene_id
+P08603	ENSG00000000971
+Q9H2S6	ENSG00000000005
+O00522	ENSG00000001631
+Q9P2G1	ENSG00000001629
+Q9BTY2	ENSG00000001036
+Q5TH74	ENSG00000001460
+P23511	ENSG00000001167
+Q6P499	ENSG00000001461
+Q9Y6X5	ENSG00000001561
+P48506	ENSG00000001084
+Q8IZE3	ENSG00000000457
+Q9NSG2	ENSG00000000460
+P09769	ENSG00000000938
+O60762	ENSG00000000419
+Q16850	ENSG00000001630
+Q9Y4W2	ENSG00000001497
+P13569	ENSG00000001626
+Q13275	ENSG00000001617
+Q1TEST	ENSG00000001617