@datagrok-libraries/bio 5.36.1 → 5.37.0

package/CHANGELOG.md

@@ -1,10 +1,17 @@
 # bio changelog
 
-## 5.37.0 (WIP)
+## 5.37.0 (2023-08-30)
+
+### Features
+
+* Added `calculateSimilarity` and `calculateIdentity` functions.
+* Added `calculateScores` function.
+
+## 5.36.1 (2023-08-18)
 
 ### Bug fixes
 
-* Restore utils getSplitter method, it is required for MacromoleculeDifference column
+* Restore utils `getSplitter` method, it is required for MacromoleculeDifference column.
 
 ## 5.36.0 (2023-08-10)
 

package/package.json

@@ -8,7 +8,7 @@
     "access": "public"
   },
   "friendlyName": "Datagrok bio library",
-  "version": "5.36.1",
+  "version": "5.37.0",
   "description": "",
   "dependencies": {
     "@datagrok-libraries/gridext": "^1.3.64",

package/src/utils/macromolecule/scoring.d.ts

@@ -0,0 +1,24 @@
+import { ISeqSplitted } from '../../utils/macromolecule/types';
+import * as DG from 'datagrok-api/dg';
+export declare enum SCORE {
+    IDENTITY = "identity",
+    SIMILARITY = "similarity"
+}
+/** Performs transformations and sequence scoring.
+ * @param {DG.DataFrame} table Table to attach results to.
+ * @param {DG.Column<string>} col Sequences column to score. Must have Macromolecule semantic type.
+ * @param {string} ref Reference sequence to score against.
+ * @param {SCORE} scoring Scoring method.
+ * @returns {DG.Column<number>} Scores column. */
+export declare function calculateScores(table: DG.DataFrame, col: DG.Column<string>, ref: string, scoring: SCORE): Promise<DG.Column<number>>;
+/** Calculates identity scores as fraction of matching monomers on the same position.
+ * @param {ISeqSplitted} reference Splitted reference sequence.
+ * @param {DG.DataFrame} positionsDf Table which only contains position columns with semantic type Monomer.
+ * @returns {DG.Column<number>} Scores column. */
+export declare function calculateIdentity(reference: ISeqSplitted, positionsDf: DG.DataFrame): DG.Column<number>;
+/** Calculates similarity scores as sum of monomer fingerprint similarities on the same position.
+ * @param {ISeqSplitted} reference Splitted reference sequence.
+ * @param {DG.DataFrame} positionsDf Table which only contains position columns with semantic type Monomer.
+ * @returns {DG.Column<number>} Scores column. */
+export declare function calculateSimilarity(reference: ISeqSplitted, positionsDf: DG.DataFrame): Promise<DG.Column<number>>;
+//# sourceMappingURL=scoring.d.ts.map

package/src/utils/macromolecule/scoring.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"scoring.d.ts","sourceRoot":"","sources":["scoring.ts"],"names":[],"mappings":"AAEA,OAAO,EAAC,YAAY,EAAC,MAAM,iCAAiC,CAAC;AAC7D,OAAO,KAAK,EAAE,MAAM,iBAAiB,CAAC;AAKtC,oBAAY,KAAK;IACf,QAAQ,aAAa;IACrB,UAAU,eAAe;CAC1B;AAED;;;;;iDAKiD;AACjD,wBAAsB,eAAe,CAAC,KAAK,EAAE,EAAE,CAAC,SAAS,EAAE,GAAG,EAAE,EAAE,CAAC,MAAM,CAAC,MAAM,CAAC,EAAE,GAAG,EAAE,MAAM,EAAE,OAAO,EAAE,KAAK,GAAG,OAAO,CAAC,EAAE,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,CAY1I;AAED;;;iDAGiD;AACjD,wBAAgB,iBAAiB,CAAC,SAAS,EAAE,YAAY,EAAE,WAAW,EAAE,EAAE,CAAC,SAAS,GAAG,EAAE,CAAC,MAAM,CAAC,MAAM,CAAC,CA0BvG;AAED;;;iDAGiD;AACjD,wBAAsB,mBAAmB,CAAC,SAAS,EAAE,YAAY,EAAE,WAAW,EAAE,EAAE,CAAC,SAAS,GAAG,OAAO,CAAC,EAAE,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,CAIxH"}

package/src/utils/macromolecule/scoring.js

@@ -0,0 +1,65 @@
+import { sequenceChemSimilarity } from '../../monomer-works/monomer-utils';
+import * as DG from 'datagrok-api/dg';
+import { splitAlignedSequences } from '../splitter';
+import { getSplitter } from './utils';
+export var SCORE;
+(function (SCORE) {
+    SCORE["IDENTITY"] = "identity";
+    SCORE["SIMILARITY"] = "similarity";
+})(SCORE || (SCORE = {}));
+/** Performs transformations and sequence scoring.
+ * @param {DG.DataFrame} table Table to attach results to.
+ * @param {DG.Column<string>} col Sequences column to score. Must have Macromolecule semantic type.
+ * @param {string} ref Reference sequence to score against.
+ * @param {SCORE} scoring Scoring method.
+ * @returns {DG.Column<number>} Scores column. */
+export async function calculateScores(table, col, ref, scoring) {
+    const splitSeqDf = splitAlignedSequences(col);
+    const splitter = getSplitter(col.getTag(DG.TAGS.UNITS), col.getTag("separator" /* bioTAGS.separator */));
+    const refSplitted = splitter(ref);
+    const scoresCol = scoring === SCORE.IDENTITY ? calculateIdentity(refSplitted, splitSeqDf) :
+        scoring === SCORE.SIMILARITY ? await calculateSimilarity(refSplitted, splitSeqDf) : null;
+    if (scoresCol === null)
+        throw new Error(`In bio library: Unkown sequence scoring method: ${scoring}`);
+    scoresCol.name = table.columns.getUnusedName(scoresCol.name);
+    table.columns.add(scoresCol);
+    return scoresCol;
+}
+/** Calculates identity scores as fraction of matching monomers on the same position.
+ * @param {ISeqSplitted} reference Splitted reference sequence.
+ * @param {DG.DataFrame} positionsDf Table which only contains position columns with semantic type Monomer.
+ * @returns {DG.Column<number>} Scores column. */
+export function calculateIdentity(reference, positionsDf) {
+    const numPositions = positionsDf.columns.length;
+    const positionCols = new Array(numPositions);
+    const positionEmptyCategories = new Array(numPositions);
+    const categoryIndexesTemplate = new Array(numPositions);
+    for (let posIdx = 0; posIdx < numPositions; ++posIdx) {
+        const posCol = positionsDf.columns.byIndex(posIdx);
+        positionCols[posIdx] = posCol.getRawData();
+        positionEmptyCategories[posIdx] = posCol.categories.indexOf('');
+        categoryIndexesTemplate[posIdx] = posCol.categories.indexOf(reference[posIdx] ?? '');
+    }
+    const identityScoresCol = DG.Column.float('Identity', positionsDf.rowCount);
+    const identityScoresData = identityScoresCol.getRawData();
+    for (let rowIndex = 0; rowIndex < positionsDf.rowCount; ++rowIndex) {
+        identityScoresData[rowIndex] = 0;
+        for (let posIdx = 0; posIdx < reference.length; ++posIdx) {
+            const categoryIndex = positionCols[posIdx][rowIndex];
+            if (categoryIndex === categoryIndexesTemplate[posIdx])
+                ++identityScoresData[rowIndex];
+        }
+        identityScoresData[rowIndex] /= reference.length;
+    }
+    return identityScoresCol;
+}
+/** Calculates similarity scores as sum of monomer fingerprint similarities on the same position.
+ * @param {ISeqSplitted} reference Splitted reference sequence.
+ * @param {DG.DataFrame} positionsDf Table which only contains position columns with semantic type Monomer.
+ * @returns {DG.Column<number>} Scores column. */
+export async function calculateSimilarity(reference, positionsDf) {
+    const monomerColumns = positionsDf.columns.toList();
+    const scoresCol = await sequenceChemSimilarity(monomerColumns, reference);
+    return scoresCol;
+}
+//# sourceMappingURL=scoring.js.map

package/src/utils/macromolecule/scoring.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"scoring.js","sourceRoot":"","sources":["scoring.ts"],"names":[],"mappings":"AACA,OAAO,EAAC,sBAAsB,EAAC,MAAM,mCAAmC,CAAC;AAEzE,OAAO,KAAK,EAAE,MAAM,iBAAiB,CAAC;AACtC,OAAO,EAAE,qBAAqB,EAAE,MAAM,aAAa,CAAC;AACpD,OAAO,EAAE,WAAW,EAAE,MAAM,SAAS,CAAC;AAGtC,MAAM,CAAN,IAAY,KAGX;AAHD,WAAY,KAAK;IACf,8BAAqB,CAAA;IACrB,kCAAyB,CAAA;AAC3B,CAAC,EAHW,KAAK,KAAL,KAAK,QAGhB;AAED;;;;;iDAKiD;AACjD,MAAM,CAAC,KAAK,UAAU,eAAe,CAAC,KAAmB,EAAE,GAAsB,EAAE,GAAW,EAAE,OAAc;IAC5G,MAAM,UAAU,GAAG,qBAAqB,CAAC,GAAG,CAAC,CAAC;IAC9C,MAAM,QAAQ,GAAG,WAAW,CAAC,GAAG,CAAC,MAAM,CAAC,EAAE,CAAC,IAAI,CAAC,KAAK,CAAC,EAAE,GAAG,CAAC,MAAM,qCAAmB,CAAC,CAAC;IACvF,MAAM,WAAW,GAAG,QAAQ,CAAC,GAAG,CAAC,CAAC;IAElC,MAAM,SAAS,GAAG,OAAO,KAAK,KAAK,CAAC,QAAQ,CAAC,CAAC,CAAC,iBAAiB,CAAC,WAAW,EAAE,UAAU,CAAC,CAAC,CAAC;QACzF,OAAO,KAAK,KAAK,CAAC,UAAU,CAAC,CAAC,CAAC,MAAM,mBAAmB,CAAC,WAAW,EAAE,UAAU,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC;IAC3F,IAAI,SAAS,KAAK,IAAI;QACpB,MAAM,IAAI,KAAK,CAAC,mDAAmD,OAAO,EAAE,CAAC,CAAC;IAChF,SAAS,CAAC,IAAI,GAAG,KAAK,CAAC,OAAO,CAAC,aAAa,CAAC,SAAS,CAAC,IAAI,CAAC,CAAC;IAC7D,KAAK,CAAC,OAAO,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;IAC7B,OAAO,SAAS,CAAC;AACnB,CAAC;AAED;;;iDAGiD;AACjD,MAAM,UAAU,iBAAiB,CAAC,SAAuB,EAAE,WAAyB;IAClF,MAAM,YAAY,GAAG,WAAW,CAAC,OAAO,CAAC,MAAM,CAAC;IAChD,MAAM,YAAY,GAAkB,IAAI,KAAK,CAAC,YAAY,CAAC,CAAC;IAC5D,MAAM,uBAAuB,GAAa,IAAI,KAAK,CAAC,YAAY,CAAC,CAAC;IAClE,MAAM,uBAAuB,GAAa,IAAI,KAAK,CAAC,YAAY,CAAC,CAAC;IAElE,KAAK,IAAI,MAAM,GAAG,CAAC,EAAE,MAAM,GAAG,YAAY,EAAE,EAAE,MAAM,EAAE;QACpD,MAAM,MAAM,GAAG,WAAW,CAAC,OAAO,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC;QACnD,YAAY,CAAC,MAAM,CAAC,GAAG,MAAM,CAAC,UAAU,EAAiB,CAAC;QAC1D,uBAAuB,CAAC,MAAM,CAAC,GAAG,MAAM,CAAC,UAAU,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC;QAChE,uBAAuB,CAAC,MAAM,CAAC,GAAG,MAAM,CAAC,UAAU,CAAC,OAAO,CAAC,SAAS,CAAC,MAAM,CAAC,IAAI,EAAE,CAAC,CAAC;KACtF;IAED,MAAM,iBAAiB,GAAG,EAAE,CAAC,MAAM,CAAC,KAAK,CAAC,UAAU,EAAE,WAAW,CAAC,QAAQ,CAAC,CAAC;IAC5E,MAAM,kBAAkB,GAAG,iBAAiB,CAAC,UAAU,EAAE,CAAC;IAC1D,KAAK,IAAI,QAAQ,GAAG,CAAC,EAAE,QAAQ,GAAG,WAAW,CAAC,QAAQ,EAAE,EAAE,QAAQ,EAAE;QAClE,kBAAkB,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC;QACjC,KAAK,IAAI,MAAM,GAAG,CAAC,EAAE,MAAM,GAAG,SAAS,CAAC,MAAM,EAAE,EAAE,MAAM,EAAE;YACxD,MAAM,aAAa,GAAG,YAAY,CAAC,MAAM,CAAC,CAAC,QAAQ,CAAC,CAAC;YACrD,IAAI,aAAa,KAAK,uBAAuB,CAAC,MAAM,CAAC;gBACnD,EAAE,kBAAkB,CAAC,QAAQ,CAAC,CAAC;SAClC;QACD,kBAAkB,CAAC,QAAQ,CAAC,IAAI,SAAS,CAAC,MAAM,CAAC;KAClD;IAED,OAAO,iBAAiB,CAAC;AAC3B,CAAC;AAED;;;iDAGiD;AACjD,MAAM,CAAC,KAAK,UAAU,mBAAmB,CAAC,SAAuB,EAAE,WAAyB;IAC1F,MAAM,cAAc,GAAG,WAAW,CAAC,OAAO,CAAC,MAAM,EAAyB,CAAC;IAC3E,MAAM,SAAS,GAAG,MAAM,sBAAsB,CAAC,cAAc,EAAE,SAAS,CAAC,CAAC;IAC1E,OAAO,SAAS,CAAC;AACnB,CAAC"}