tooluniverse 1.0.10__py3-none-any.whl → 1.0.11__py3-none-any.whl

tooluniverse/data/geo_tools.json

@@ -1,49 +1,39 @@
 [
   {
-    "type": "GEORESTTool",
-    "name": "GEO_search_expression_data",
-    "description": "Search gene expression data from the GEO database. GEO is a public repository that archives and freely distributes microarray, next-generation sequencing, and other forms of high-throughput functional genomics data.",
+    "type": "GEOSearchDatasets",
+    "name": "geo_search_datasets",
+    "description": "Search GEO datasets by query terms, organism, study type, or platform. Returns dataset IDs and basic information.",
     "parameter": {
       "type": "object",
       "properties": {
         "query": {
           "type": "string",
-          "description": "Search query (e.g., 'cancer', 'diabetes', 'microarray')",
-          "minLength": 1
+          "description": "Search query terms (e.g., 'cancer', 'diabetes', 'microarray')"
         },
         "organism": {
           "type": "string",
           "description": "Organism name (e.g., 'Homo sapiens', 'Mus musculus')",
-          "default": "Homo sapiens"
+          "default": ""
         },
         "study_type": {
           "type": "string",
-          "description": "Type of study (e.g., 'expression', 'methylation', 'genome')",
-          "enum": ["expression", "methylation", "genome", "sequence", "other"]
+          "description": "Type of study (e.g., 'Expression profiling by array', 'Expression profiling by high throughput sequencing')",
+          "default": ""
         },
         "platform": {
           "type": "string",
-          "description": "Platform used (e.g., 'GPL96', 'GPL570')"
-        },
-        "date_range": {
-          "type": "string",
-          "description": "Date range in format 'YYYY:YYYY' (e.g., '2020:2023')"
+          "description": "Platform used (e.g., 'GPL570', 'GPL96')",
+          "default": ""
         },
         "limit": {
           "type": "integer",
-          "description": "Maximum number of results to return (default: 50)",
+          "description": "Maximum number of results to return",
+          "default": 50,
           "minimum": 1,
-          "maximum": 500,
-          "default": 50
-        },
-        "sort": {
-          "type": "string",
-          "description": "Sort order ('relevance', 'date', 'title')",
-          "enum": ["relevance", "date", "title"],
-          "default": "relevance"
+          "maximum": 500
         }
       },
-      "required": ["query"]
+      "required": []
     },
     "fields": {
       "endpoint": "/esearch.fcgi",
@@ -51,36 +41,198 @@
     },
     "return_schema": {
       "type": "object",
+      "description": "GEO dataset search results",
       "properties": {
-        "success": {"type": "boolean"},
-        "count": {"type": "integer"},
-        "studies": {
-          "type": "array",
-          "items": {
-            "type": "object",
-            "properties": {
-              "id": {"type": "string"},
-              "title": {"type": "string"},
-              "summary": {"type": "string"},
-              "organism": {"type": "string"},
-              "platform": {"type": "string"},
-              "samples": {"type": "integer"},
-              "series_type": {"type": "string"},
-              "publication_date": {"type": "string"},
-              "submission_date": {"type": "string"},
-              "contact": {"type": "string"},
-              "citation": {"type": "string"}
+        "esearchresult": {
+          "type": "object",
+          "properties": {
+            "count": {
+              "type": "string",
+              "description": "Number of datasets found"
+            },
+            "idlist": {
+              "type": "array",
+              "items": {
+                "type": "string"
+              },
+              "description": "List of GEO dataset IDs"
+            },
+            "querytranslation": {
+              "type": "string",
+              "description": "Translated search query"
             }
           }
-        },
-        "query_translation": {"type": "string"},
-        "error": {"type": "string"}
+        }
+      }
+    },
+    "test_examples": [
+      {
+        "query": "cancer",
+        "organism": "Homo sapiens",
+        "limit": 10
+      },
+      {
+        "query": "diabetes",
+        "study_type": "Expression profiling by array",
+        "limit": 5
+      }
+    ],
+    "label": [
+      "GEO",
+      "Dataset",
+      "Search"
+    ],
+    "metadata": {
+      "tags": ["expression", "microarray", "search"],
+      "difficulty_level": "easy",
+      "estimated_execution_time": "< 3 seconds"
+    }
+  },
+  {
+    "type": "GEOGetDatasetInfo",
+    "name": "geo_get_dataset_info",
+    "description": "Get detailed information about a specific GEO dataset including title, summary, and metadata.",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "dataset_id": {
+          "type": "string",
+          "description": "GEO dataset ID (e.g., 'GDS1234', 'GSE12345')"
+        }
+      },
+      "required": ["dataset_id"]
+    },
+    "fields": {
+      "endpoint": "/esummary.fcgi",
+      "return_format": "JSON"
+    },
+    "return_schema": {
+      "type": "object",
+      "description": "GEO dataset information",
+      "properties": {
+        "result": {
+          "type": "object",
+          "description": "Dataset information from GEO",
+          "properties": {
+            "uid": {
+              "type": "string",
+              "description": "Dataset UID"
+            },
+            "title": {
+              "type": "string",
+              "description": "Dataset title"
+            },
+            "summary": {
+              "type": "string",
+              "description": "Dataset summary"
+            },
+            "organism": {
+              "type": "string",
+              "description": "Organism"
+            },
+            "platform": {
+              "type": "string",
+              "description": "Platform used"
+            },
+            "samples": {
+              "type": "string",
+              "description": "Number of samples"
+            }
+          }
+        }
+      }
+    },
+    "test_examples": [
+      {
+        "dataset_id": "GDS1234"
+      },
+      {
+        "dataset_id": "GSE12345"
+      }
+    ],
+    "label": [
+      "GEO",
+      "Dataset",
+      "Information"
+    ],
+    "metadata": {
+      "tags": ["dataset", "metadata", "expression"],
+      "difficulty_level": "easy",
+      "estimated_execution_time": "< 3 seconds"
+    }
+  },
+  {
+    "type": "GEOGetSampleInfo",
+    "name": "geo_get_sample_info",
+    "description": "Get sample information for a GEO dataset including sample characteristics and experimental conditions.",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "dataset_id": {
+          "type": "string",
+          "description": "GEO dataset ID (e.g., 'GDS1234', 'GSE12345')"
+        }
+      },
+      "required": ["dataset_id"]
+    },
+    "fields": {
+      "endpoint": "/esummary.fcgi",
+      "return_format": "JSON"
+    },
+    "return_schema": {
+      "type": "object",
+      "description": "GEO sample information",
+      "properties": {
+        "result": {
+          "type": "object",
+          "description": "Sample information from GEO",
+          "properties": {
+            "uid": {
+              "type": "string",
+              "description": "Dataset UID"
+            },
+            "samples": {
+              "type": "array",
+              "description": "Array of sample information",
+              "items": {
+                "type": "object",
+                "properties": {
+                  "sample_id": {
+                    "type": "string",
+                    "description": "Sample identifier"
+                  },
+                  "characteristics": {
+                    "type": "string",
+                    "description": "Sample characteristics"
+                  },
+                  "treatment": {
+                    "type": "string",
+                    "description": "Treatment conditions"
+                  }
+                }
+              }
+            }
+          }
+        }
       }
     },
-    "implementation": {
-      "language": "python",
-      "dependencies": ["requests"],
-      "source_file": "geo_tool.py"
+    "test_examples": [
+      {
+        "dataset_id": "GDS1234"
+      },
+      {
+        "dataset_id": "GSE12345"
+      }
+    ],
+    "label": [
+      "GEO",
+      "Sample",
+      "Information"
+    ],
+    "metadata": {
+      "tags": ["sample", "characteristics", "expression"],
+      "difficulty_level": "easy",
+      "estimated_execution_time": "< 3 seconds"
     }
   }
 ]

tooluniverse/data/gnomad_tools.json

@@ -0,0 +1,109 @@
+[
+  {
+    "type": "gnomADGetGeneConstraints",
+    "name": "gnomad_get_gene_constraints",
+    "description": "Get gene constraint metrics from gnomAD. Returns pLI, LOEUF, and other constraint scores for genes.",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "gene_symbol": {
+          "type": "string",
+          "description": "Gene symbol (e.g., 'BRCA1', 'TP53')"
+        }
+      },
+      "required": ["gene_symbol"]
+    },
+    "fields": {
+      "endpoint": "https://gnomad.broadinstitute.org/api",
+      "return_format": "JSON",
+      "query_schema": "query GeneConstraints($geneSymbol: String!) { gene(gene_symbol: $geneSymbol, reference_genome: GRCh38) { symbol gene_id exac_constraint { exp_lof obs_lof pLI exp_mis obs_mis exp_syn obs_syn } gnomad_constraint { exp_lof obs_lof oe_lof pLI exp_mis obs_mis oe_mis exp_syn obs_syn oe_syn } } }"
+    },
+    "return_schema": {
+      "type": "object",
+      "description": "gnomAD gene constraint data",
+      "properties": {
+        "status": {
+          "type": "string",
+          "description": "Status of the request"
+        },
+        "data": {
+          "type": "object",
+          "description": "GraphQL response data",
+          "properties": {
+            "data": {
+              "type": "object",
+              "properties": {
+                "gene": {
+                  "type": "object",
+                  "properties": {
+                    "symbol": {
+                      "type": "string",
+                      "description": "Gene symbol"
+                    },
+                    "gene_id": {
+                      "type": "string",
+                      "description": "Ensembl gene ID"
+                    },
+                    "exac_constraint": {
+                      "type": "object",
+                      "description": "ExAC constraint metrics",
+                      "properties": {
+                        "exp_lof": {
+                          "type": "number",
+                          "description": "Expected number of loss-of-function variants"
+                        },
+                        "obs_lof": {
+                          "type": "number",
+                          "description": "Observed number of loss-of-function variants"
+                        },
+                        "pLI": {
+                          "type": "number",
+                          "description": "Probability of being loss-of-function intolerant"
+                        }
+                      }
+                    },
+                    "gnomad_constraint": {
+                      "type": "object",
+                      "description": "gnomAD constraint metrics",
+                      "properties": {
+                        "oe_lof": {
+                          "type": "number",
+                          "description": "Observed/Expected ratio for loss-of-function variants"
+                        },
+                        "pLI": {
+                          "type": "number",
+                          "description": "Probability of being loss-of-function intolerant"
+                        }
+                      }
+                    }
+                  }
+                }
+              }
+            }
+          }
+        },
+        "gene_symbol": {
+          "type": "string",
+          "description": "Input gene symbol"
+        }
+      }
+    },
+    "test_examples": [
+      {
+        "description": "Get constraint metrics for BRCA1 gene",
+        "input": {
+          "gene_symbol": "BRCA1"
+        },
+        "expected_output": {
+          "status": "success",
+          "data": "Gene constraint data with pLI and LOEUF scores"
+        }
+      }
+    ],
+    "metadata": {
+      "tags": ["constraint", "pLI", "LOEUF", "gnomAD"],
+      "difficulty_level": "intermediate",
+      "estimated_execution_time": "< 3 seconds"
+    }
+  }
+]

tooluniverse/data/gtopdb_tools.json

@@ -0,0 +1,68 @@
+[
+  {
+    "type": "GtoPdbRESTTool",
+    "name": "GtoPdb_get_targets",
+    "description": "Get pharmacological targets from Guide to Pharmacology database",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "target_type": {"type": "string", "default": "protein", "description": "Target type"},
+        "limit": {"type": "integer", "default": 20, "description": "Number of results"}
+      }
+    },
+    "fields": {
+      "endpoint": "https://www.guidetopharmacology.org/services/targets",
+      "return_format": "JSON"
+    },
+    "return_schema": {
+      "type": "array",
+      "items": {
+        "type": "object",
+        "properties": {
+          "targetId": {"type": "integer"},
+          "name": {"type": "string"},
+          "targetType": {"type": "string"},
+          "species": {"type": "string"},
+          "uniprotId": {"type": "string"},
+          "ensemblGeneId": {"type": "string"},
+          "hgncId": {"type": "string"},
+          "description": {"type": "string"},
+          "family": {"type": "string"},
+          "subfamily": {"type": "string"},
+          "subunit": {"type": "string"},
+          "tissue": {"type": "array"},
+          "cellLine": {"type": "array"},
+          "diseases": {"type": "array"},
+          "drugs": {
+            "type": "array",
+            "items": {
+              "type": "object",
+              "properties": {
+                "drugId": {"type": "integer"},
+                "name": {"type": "string"},
+                "action": {"type": "string"},
+                "mechanism": {"type": "string"}
+              }
+            }
+          },
+          "ligands": {
+            "type": "array",
+            "items": {
+              "type": "object",
+              "properties": {
+                "ligandId": {"type": "integer"},
+                "name": {"type": "string"},
+                "type": {"type": "string"},
+                "affinity": {"type": "string"}
+              }
+            }
+          }
+        }
+      }
+    },
+    "test_examples": [
+      {"target_type": "protein", "limit": 5},
+      {"target_type": "receptor", "limit": 10}
+    ]
+  }
+]

tooluniverse/data/gwas_tools.json

@@ -1839,5 +1839,37 @@
         }
       }
     }
+  },
+  {
+    "type": "GWASGeneSearch",
+    "name": "GWAS_search_associations_by_gene",
+    "description": "Search GWAS Catalog associations by gene name (returns strongest risk allele and p-value fields).",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "gene_name": {"type": "string", "description": "Gene symbol (e.g., BRCA1)."},
+        "size": {"type": "integer", "description": "Max associations to return.", "default": 5}
+      },
+      "required": ["gene_name"]
+    },
+    "return_schema": {
+      "type": "object",
+      "properties": {
+        "gene_name": {"type": "string"},
+        "association_count": {"type": "integer"},
+        "associations": {
+          "type": "array",
+          "items": {
+            "type": "object",
+            "properties": {
+              "study": {"type": "string"},
+              "strongestRiskAllele": {"type": "string"},
+              "pvalue": {"type": "number"}
+            }
+          }
+        },
+        "total_found": {"type": "integer"}
+      }
+    }
   }
 ]