Merge pull request #3 from mite-standard/dev_mmz

Dev mmz
mite-standard · Aug 5, 2024 · 897fb77 · 897fb77
2 parents 5360a0d + 1489387
commit 897fb77
Show file tree

Hide file tree

Showing 9 changed files with 142 additions and 61 deletions.
diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml
@@ -0,0 +1,51 @@
+name: Publish Python distribution package to PyPI
+
+on:
+  release:
+    types: [published]
+
+jobs:
+  build:
+    name: Build distribution package
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v4
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.x"
+      - name: Install pypa/build
+        run: >-
+          python3 -m
+          pip install
+          build
+          --user
+      - name: Build a binary wheel and a source tarball
+        run: python3 -m build
+      - name: Store the distribution packages
+        uses: actions/upload-artifact@v4
+        with:
+          name: python-package-distributions
+          path: dist/
+
+  publish-to-pypi:
+    name: >-
+      Publish Python distribution package to PyPI
+    needs:
+      - build
+    runs-on: ubuntu-latest
+    environment:
+      name: pypi
+      url: https://pypi.org/p/mite_schema
+    permissions:
+      id-token: write
+
+    steps:
+      - name: Download all the dists
+        uses: actions/download-artifact@v4
+        with:
+          name: python-package-distributions
+          path: dist/
+      - name: Publish distribution package to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -5,11 +5,18 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.1] 05-08-2024
+
+### Changed
+
+- schema/enzyme: 'databaseIds' is now an object, with allowed database cross-links as keys (before: an array of database-crosslinks)
+- schema/reactions: 'databaseIds' is now an object, with allowed database cross-links as keys (before: an array of database-crosslinks)
+
 ## [1.0] 05-08-2024
 
 ### Added
 
-- Implemented MITE Schmea and SchemaManager()
+- Implemented MITE Schema and SchemaManager()
 
 ### Changed
 

diff --git a/README.md b/README.md
@@ -14,11 +14,12 @@ More information about the schema can be found in [the MITE publication](https:/
 
 ## Quick Start
 
+To validate your MITE-formatted .json-file, run:
+
 - `hatch run mite_schema -i <input.json>`
 
 ### For developers
 
-- Install `python3`
 - Install `hatch` using one of the methods described [here](https://hatch.pypa.io/1.12/install/)
 - Download or clone this repository
 - Run `hatch -v env create dev`. This will download and install the appropriate Python version and any required packages

diff --git a/mite_schema/schema/definitions/enzyme.json b/mite_schema/schema/definitions/enzyme.json
@@ -17,11 +17,19 @@
       "type": "string"
     },
     "databaseIds": {
-      "type": "array",
-      "items": {
-        "$ref": "#/$defs/databaseId"
-      },
-      "uniqueItems": true
+      "type": "object",
+      "additionalProperties": false,
+      "properties": {
+        "uniprot": {
+          "$ref": "#/$defs/uniprot"
+        },
+        "genpept": {
+          "$ref": "#/$defs/genpept"
+        },
+        "mibig": {
+          "$ref": "#/$defs/mibig"
+        }
+      }
     },
     "auxiliaryEnzymes": {
       "title": "Specify any required auxiliary enzymes that are co-forming the maturation machinery.",
@@ -39,11 +47,16 @@
             "type": "string"
           },
           "databaseIds": {
-            "type": "array",
-            "items": {
-              "$ref": "#/$defs/databaseId"
-            },
-            "uniqueItems": true
+            "type": "object",
+            "additionalProperties": false,
+            "properties": {
+              "uniprot": {
+                "$ref": "#/$defs/uniprot"
+              },
+              "genpept": {
+                "$ref": "#/$defs/genpept"
+              }
+            }
           }
         }
       }
@@ -58,26 +71,20 @@
     }
   },
   "$defs": {
-    "databaseId": {
+    "uniprot": {
+      "title": "Uniprot ID reference.",
       "type": "string",
-      "oneOf": [
-        {
-          "title": "Uniprot ID reference.",
-          "pattern": "^uniprot:[A-Z0-9]+$"
-        },
-        {
-          "title": "GenBank GenPept ID (= protein ID, GenBank gene products)",
-          "pattern": "^genpept:[A-Z]{3}[0-9]{5,7}\\.[0-9]+$"
-        },
-        {
-          "title": "MIBiG ID reference",
-          "pattern": "^mibig:BGC\\d{7}$"
-        },
-        {
-          "title": "Temporary MIBiG ID reference",
-          "pattern": "^mibig:new\\d+$"
-        }
-      ]
+      "pattern": "^[A-Z0-9]+$"
+    },
+    "genpept": {
+      "title": "GenBank GenPept ID (= protein ID, GenBank gene products)",
+      "type": "string",
+      "pattern": "^[A-Z]{3}[0-9]{5,7}\\.[0-9]+$"
+    },
+    "mibig": {
+      "title": "MIBiG ID reference",
+      "type": "string",
+      "pattern": "^BGC\\d{7}$|^new\\d+$"
     }
   }
 }
diff --git a/mite_schema/schema/definitions/reactions.json b/mite_schema/schema/definitions/reactions.json
@@ -30,11 +30,19 @@
       "uniqueItems": true
     },
     "databaseIds": {
-      "type": "array",
-      "items": {
-        "$ref": "#/$defs/databaseId"
-      },
-      "uniqueItems": true
+      "type": "object",
+      "additionalProperties": false,
+      "properties": {
+        "rhea": {
+          "$ref": "#/$defs/rhea"
+        },
+        "mite": {
+          "$ref": "#/$defs/mite"
+        },
+        "ec": {
+          "$ref": "#/$defs/ec"
+        }
+      }
     },
     "reactionSMARTS": {
       "title": "A reaction SMARTS pattern describing the (sub)structure to match and the reaction to perform.",
@@ -139,20 +147,20 @@
         "Other"
       ]
     },
-    "databaseId": {
-      "title": "Cross-reference to databases.",
+    "rhea": {
+      "title": "RHEA cross-reference",
       "type": "string",
-      "oneOf": [
-        {
-          "pattern": "^rhea:(\\d+)$"
-        },
-        {
-          "pattern": "^MITE(\\d{7,7})$"
-        },
-        {
-          "pattern": "^EC [0-9]+(\\.[0-9]+){0,3}$"
-        }
-      ]
+      "pattern": "^(\\d+)$"
+    },
+    "mite": {
+      "title": "MITE cross-reference",
+      "type": "string",
+      "pattern": "^MITE(\\d{7})$"
+    },
+    "ec": {
+      "title": "EC number (2-4 identifiers)",
+      "type": "string",
+      "pattern": "^EC [0-9]+(\\.[0-9]+){0,3}$"
     },
     "evidence": {
       "type": "object",

diff --git a/mite_schema/schema_manager/__init__.py b/mite_schema/schema_manager/__init__.py
diff --git a/mite_schema/schema_manager/schema_manager.py b/mite_schema/schema_manager/schema_manager.py
@@ -148,5 +148,6 @@ def validate_mite(self: Self, instance: dict):
             jsonschema.validate(instance=instance, schema=entry, registry=registry)
         except jsonschema.exceptions.ValidationError as e:
             raise ValueError(
-                f"SchemaManager: Validation of instance against MITE schema led to an error: '{e!s}"
+                f"SchemaManager: Validation of instance against "
+                f"MITE schema led to an error: '{e!s}"
             ) from e
diff --git a/pyproject.toml b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "mite_schema"
-version = "1.0"
+version = "1.1"
 description = "Containing the Minimum Information about a Tailoring Enzymes schema and auxiliary methods"
 readme = "README.md"
 requires-python = ">=3.12"
@@ -24,7 +24,6 @@ classifiers = [
     "Topic :: Scientific/Engineering :: Bio-Informatics",
     "Topic :: Scientific/Engineering :: Chemistry"
 ]
-
 dependencies = [
     "argparse~=1.4",
     "coloredlogs~=15.0",
@@ -33,6 +32,10 @@ dependencies = [
     "referencing~=0.35",
 ]
 
+[project.urls]
+Homepage = "https://github.com/mite-standard/mite_schema"
+Issues = "https://github.com/mite-standard/mite_schema/issues"
+
 [project.scripts]
 mite_schema = "mite_schema.main:main"
 

diff --git a/tests/example_files/example_valid.json b/tests/example_files/example_valid.json
@@ -26,11 +26,11 @@
     "enzyme": {
         "name": "McjC",
         "description": "Lassopeptide ATP-dependent lactam synthetase (C-protein, asparagine synthetase-like)",
-        "databaseIds": [
-            "uniprot:Q9X2V9",
-            "genpept:AAD28496.1",
-            "mibig:BGC0000581"
-        ],
+        "databaseIds": {
+            "uniprot": "Q9X2V9",
+            "genpept": "AAD28496.1",
+            "mibig": "BGC0000581"
+        },
         "references": [
             "doi:10.1128/jb.181.8.2659-2662.1999",
             "doi:10.1002/cbic.201200016"
@@ -39,10 +39,10 @@
             {
                 "name": "McjB",
                 "description": "Lassopeptide cysteine protease. Required for macrolactam-synthethase function (folds and cleaves precursor peptide)",
-                "databaseIds": [
-                    "uniprot:Q9X2V8",
-                    "genpept:AAD28495.1"
-                ]
+                "databaseIds": {
+                    "uniprot": "Q9X2V8",
+                    "genpept": "AAD28495.1"
+                }
             }
         ]
     },
@@ -67,6 +67,9 @@
                     "description": "Lasso macrocyclisation of microcin J25 after precursor cleavage, leading to mature product."
                 }
             ],
+            "databaseIds": {
+                "mite": "MITE0000000"
+            },
             "evidence": [
                 {
                     "evidenceCode": [