@datagrok/peptides 1.27.6 → 1.27.8

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@datagrok/peptides",
   "friendlyName": "Peptides",
-  "version": "1.27.6",
+  "version": "1.27.8",
   "author": {
     "name": "Davit Rizhinashvili",
     "email": "drizhinashvili@datagrok.ai"
@@ -13,7 +13,7 @@
     "directory": "packages/Peptides"
   },
   "dependencies": {
-    "@datagrok-libraries/bio": "^5.63.2",
+    "@datagrok-libraries/bio": "^5.63.6",
     "@datagrok-libraries/math": "^1.2.6",
     "@datagrok-libraries/ml": "^6.10.7",
     "@datagrok-libraries/statistics": "^1.2.12",
@@ -31,11 +31,11 @@
   "devDependencies": {
     "@datagrok-libraries/helm-web-editor": "^1.1.16",
     "@datagrok-libraries/js-draw-lite": "^0.0.5",
-    "@datagrok/bio": "^2.24.0",
-    "@datagrok/chem": "^1.13.0",
+    "@datagrok/bio": "^2.26.4",
+    "@datagrok/chem": "^1.17.1",
     "@datagrok/dendrogram": "^1.2.33",
     "@datagrok/eda": "^1.3.1",
-    "@datagrok/helm": "^2.7.0",
+    "@datagrok/helm": "^2.13.0",
     "@types/uuid": "^10.0.0",
     "@types/wu": "^2.1.44",
     "@typescript-eslint/eslint-plugin": "^8.8.1",

package/src/model.ts

@@ -50,6 +50,7 @@ import {ITSNEOptions, IUMAPOptions}
 import {DimReductionMethods} from '@datagrok-libraries/ml/src/multi-column-dimensionality-reduction/types';
 import {AggregationColumns, MonomerPositionStats} from './utils/statistics';
 import {splitAlignedSequences} from '@datagrok-libraries/bio/src/utils/splitter';
+import {MONOMER_CANONICALIZER_FUNC_TAG} from '@datagrok-libraries/bio/src/utils/macromolecule/consts';
 import {getDbscanWorker} from '@datagrok-libraries/math';
 import {DistanceAggregationMethods} from '@datagrok-libraries/ml/src/distance-matrix/types';
 import {ClusterMaxActivityViewer, IClusterMaxActivity} from './viewers/cluster-max-activity-viewer';
@@ -748,9 +749,14 @@ export class PeptidesModel {
     const positionColumns = splitSeqDf.columns.names();
     if (positionColumns.every((colName) => cols.contains(colName))) {
       positionColumns.forEach((colName) => {
-        this.df.col(colName)!.setTag(C.TAGS.ANALYSIS_COL, `${true}`);
-        this.df.col(colName)!.setTag(C.TAGS.POSITION_COL, `${true}`);
-        CR.setMonomerRenderer(this.df.col(colName)!, this.alphabet);
+        const col = this.df.col(colName)!;
+        col.setTag(C.TAGS.ANALYSIS_COL, `${true}`);
+        col.setTag(C.TAGS.POSITION_COL, `${true}`);
+        // Propagate canonicalizer tag from split columns if present
+        const canonTag = splitSeqDf.getCol(colName).getTag(MONOMER_CANONICALIZER_FUNC_TAG);
+        if (canonTag)
+          col.setTag(MONOMER_CANONICALIZER_FUNC_TAG, canonTag);
+        CR.setMonomerRenderer(col, this.alphabet);
       });
     } else {
       for (const colName of positionColumns) {
@@ -763,6 +769,10 @@ export class PeptidesModel {
         col = cols.addNew(newCol.name, newCol.type).init((i) => newColCat[newColData[i]]);
         col.setTag(C.TAGS.ANALYSIS_COL, `${true}`);
         col.setTag(C.TAGS.POSITION_COL, `${true}`);
+        // Propagate canonicalizer tag from split columns if present
+        const canonTag = newCol.getTag(MONOMER_CANONICALIZER_FUNC_TAG);
+        if (canonTag)
+          col.setTag(MONOMER_CANONICALIZER_FUNC_TAG, canonTag);
         CR.setMonomerRenderer(col, this.alphabet);
       }
     }

package/src/viewers/mutation-cliffs-viewer.ts

@@ -10,6 +10,8 @@ import $ from 'cash-dom';
 import {PeptidesModel} from '../model';
 import {extractColInfo} from '../utils/misc';
 import {Subscription} from 'rxjs';
+import {SeqTemps} from '@datagrok-libraries/bio/src/utils/macromolecule/seq-handler';
+import {MONOMER_CANONICALIZER_FUNC_TAG} from '@datagrok-libraries/bio/src/utils/macromolecule/consts';
 export type MutationCliffsWithMonomers = {
   cliffs: MutationCliffs,
   monomers: string[]
@@ -267,15 +269,20 @@ export class MutationCliffsViewer extends DG.JsViewer {
       this._positionColumns = posCols;
       return this._positionColumns;
     }
-    // fallback: generate columns
+    // fallback: generate columns using original (non-canonical) monomers
     const seqCol = this.dataFrame.col(this.sequenceColumnName)!;
     const seqHelper = PeptideUtils.getSeqHelper();
     const seqHandler = seqHelper.getSeqHandler(seqCol);
     const length = seqHandler.maxLength;
     const cols: DG.Column[] = [];
+    // Propagate canonicalizer tag if the source column has a notation provider with one
+    const notationProvider: any = seqCol.temp[SeqTemps.notationProvider] ?? null;
+    const canonFuncName: string | null = notationProvider?.monomerCanonicalizerFuncName ?? null;
     for (let i = 0; i < length; i++) {
-      const monomersAtPosition = seqHandler.getMonomersAtPosition(i, true);
+      const monomersAtPosition = seqHandler.getMonomersAtPosition(i, false);
       const monomerCol = DG.Column.fromList('string', `Position ${i + 1}`, monomersAtPosition);
+      if (canonFuncName)
+        monomerCol.setTag(MONOMER_CANONICALIZER_FUNC_TAG, canonFuncName);
       cols.push(monomerCol);
     }
     this._positionColumns = cols;

package/src/viewers/position-statistics-viewer.ts

@@ -93,8 +93,8 @@ export class SequencePositionStatsViewer extends DG.JsViewer {
     const seqHandler = seqHelper.getSeqHandler(sequenceColumn);
     const maxPos = seqHandler.maxLength;
 
-    const canonicals = positions.map((p) => seqHandler.getMonomersAtPosition(p - 1, true));
-    this._positionColumn.init((i) => canonicals.map((c) => c[i]).join(MONOMER_MOTIF_SPLITTER));
+    const originals = positions.map((p) => seqHandler.getMonomersAtPosition(p - 1, false));
+    this._positionColumn.init((i) => originals.map((c) => c[i]).join(MONOMER_MOTIF_SPLITTER));
 
     this._boxPlotViewer = this.dataFrame.plot.box({categoryColumnNames: [this._positionColumn.name], plotStyle: 'violin',
       valueColumnName: this.valueColumnName, colorColumnName: this._positionColumn.name, showColorSelector: false, showSizeSelector: false, showCategorySelector: false,

package/src/viewers/sar-viewer.ts

@@ -36,12 +36,14 @@ import {
 } from '../utils/misc';
 import {splitAlignedSequences} from '@datagrok-libraries/bio/src/utils/splitter';
 import {LogoSummaryTable} from './logo-summary';
-import {TAGS as bioTAGS} from '@datagrok-libraries/bio/src/utils/macromolecule/consts';
+import {TAGS as bioTAGS, MONOMER_CANONICALIZER_FUNC_TAG} from '@datagrok-libraries/bio/src/utils/macromolecule/consts';
 import {ALPHABET} from '@datagrok-libraries/bio/src/utils/macromolecule';
 import {getMonomerLibHelper} from '@datagrok-libraries/bio/src/types/monomer-library';
 import {PolymerTypes} from '@datagrok-libraries/bio/src/helm/consts';
 import {PeptideUtils} from '../peptideUtils';
 import {StringDictionary} from '@datagrok-libraries/utils/src/type-declarations';
+import {SeqTemps} from '@datagrok-libraries/bio/src/utils/macromolecule/seq-handler';
+import {getSeparator} from '../utils/misc';
 
 export enum SELECTION_MODE {
   MUTATION_CLIFFS = 'Mutation Cliffs',
@@ -244,6 +246,16 @@ export abstract class SARViewer extends DG.JsViewer implements ISARViewer {
     return this._positionColumns!;
   }
 
+  /** Copies the monomer canonicalizer tag from the sequence column to a monomer column. */
+  protected propagateCanonicalizerTag(monomerCol: DG.Column): void {
+    const seqCol = this.dataFrame?.col(this.sequenceColumnName);
+    if (!seqCol) return;
+    const notationProvider: any = seqCol.temp[SeqTemps.notationProvider] ?? null;
+    const canonFuncName: string | null = notationProvider?.monomerCanonicalizerFuncName ?? null;
+    if (canonFuncName)
+      monomerCol.setTag(MONOMER_CANONICALIZER_FUNC_TAG, canonFuncName);
+  }
+
   _monomerPositionStats: MonomerPositionStats | null = null;
 
   /**
@@ -467,7 +479,8 @@ export abstract class SARViewer extends DG.JsViewer implements ISARViewer {
    */
   onPropertyChanged(property: DG.Property): void {
     super.onPropertyChanged(property);
-
+    if (!this.dataFrame)
+      return;
 
     this.doRender = true;
     switch (property.name) {
@@ -543,6 +556,152 @@ export abstract class SARViewer extends DG.JsViewer implements ISARViewer {
     throw new Error('Not implemented');
   }
 
+  /** Shows a dialog to choose extra columns, then exports all unique mutation cliffs as a new table view. */
+  exportMutationCliffs(): void {
+    if (!this.mutationCliffs) {
+      grok.shell.warning('Mutation cliffs have not been calculated yet.');
+      return;
+    }
+
+    const available = this.dataFrame.columns.toList()
+      .filter((col) => col.name !== this.activityColumnName && col.name !== this.sequenceColumnName &&
+        !this.positionColumns.some((pc) => pc.name === col.name))
+      .map((col) => col.name);
+
+    const columnsInput = ui.input.columns('Extra columns', {table: this.dataFrame, value: [], available, nullable: true});
+
+    ui.dialog('Export Mutation Cliffs')
+      .add(columnsInput.root)
+      .onOK(() => this._doExportMutationCliffs(columnsInput.value ?? []))
+      .show();
+  }
+
+  private _doExportMutationCliffs(extraColumns: DG.Column[]): void {
+    const mc = this.mutationCliffs!;
+    const alignedSeqCol = this.dataFrame.getCol(this.sequenceColumnName);
+    const alignedSeqColCategories = alignedSeqCol.categories;
+    const alignedSeqColData = alignedSeqCol.getRawData();
+    const activityCol = this.dataFrame.getCol(this.activityColumnName);
+    const activityColData = activityCol.getRawData();
+
+    const seq1Array: string[] = [];
+    const seq2Array: string[] = [];
+    const diffArray: string[] = [];
+    const act1Array: (number | null)[] = [];
+    const act2Array: (number | null)[] = [];
+    const deltaArray: (number | null)[] = [];
+    const extraData1: any[][] = extraColumns.map(() => []);
+    const extraData2: any[][] = extraColumns.map(() => []);
+
+    const seenPairs = new Set<string>();
+
+    for (const [_monomer, posMap] of mc.entries()) {
+      for (const [_position, indexMap] of posMap.entries()) {
+        for (const [refIdx, subIndexes] of indexMap.entries()) {
+          for (const subIdx of subIndexes) {
+            const pairKey = refIdx < subIdx ? `${refIdx}-${subIdx}` : `${subIdx}-${refIdx}`;
+            if (seenPairs.has(pairKey))
+              continue;
+            seenPairs.add(pairKey);
+
+            const seq1 = alignedSeqColCategories[alignedSeqColData[refIdx]];
+            const seq2 = alignedSeqColCategories[alignedSeqColData[subIdx]];
+            seq1Array.push(seq1);
+            seq2Array.push(seq2);
+            diffArray.push(`${seq1}#${seq2}`);
+
+            const a1 = activityCol.isNone(refIdx) ? null : activityColData[refIdx];
+            const a2 = activityCol.isNone(subIdx) ? null : activityColData[subIdx];
+            act1Array.push(a1);
+            act2Array.push(a2);
+            deltaArray.push(a1 == null || a2 == null ? null : a1 - a2);
+
+            for (let e = 0; e < extraColumns.length; e++) {
+              const eCol = extraColumns[e];
+              extraData1[e].push(eCol.isNone(refIdx) ? null : eCol.get(refIdx));
+              extraData2[e].push(eCol.isNone(subIdx) ? null : eCol.get(subIdx));
+            }
+          }
+        }
+      }
+    }
+
+    if (seq1Array.length === 0) {
+      grok.shell.warning('No mutation cliffs found to export.');
+      return;
+    }
+
+    const seq1Col = DG.Column.fromStrings('Seq 1', seq1Array);
+    const seq2Col = DG.Column.fromStrings('Seq 2', seq2Array);
+    const diffCol = DG.Column.fromStrings('Mutation', diffArray);
+
+    // Copy sequence tags (without notation provider) so the platform detects the same semtype
+    for (const col of [seq1Col, seq2Col]) {
+      for (const tag of alignedSeqCol.tags.keys()) {
+        if (tag !== '.notationProvider')
+          col.setTag(tag, alignedSeqCol.getTag(tag)!);
+      }
+      col.semType = alignedSeqCol.semType;
+    }
+
+    // Set up macromolecule difference column
+    diffCol.semType = C.SEM_TYPES.MACROMOLECULE_DIFFERENCE;
+    diffCol.setTag(C.TAGS.SEPARATOR, getSeparator(alignedSeqCol));
+    diffCol.setTag(DG.TAGS.UNITS, alignedSeqCol.getTag(DG.TAGS.UNITS) ?? '');
+    diffCol.setTag(DG.TAGS.CELL_RENDERER, 'MacromoleculeDifference');
+    diffCol.temp[SeqTemps.notationProvider] = alignedSeqCol.temp[SeqTemps.notationProvider];
+
+    const act1Col = DG.Column.fromList(DG.COLUMN_TYPE.FLOAT, `Seq 1 ${this.activityColumnName}`, act1Array);
+    const act2Col = DG.Column.fromList(DG.COLUMN_TYPE.FLOAT, `Seq 2 ${this.activityColumnName}`, act2Array);
+    const deltaCol = DG.Column.fromList(DG.COLUMN_TYPE.FLOAT, 'Delta', deltaArray);
+
+    const columns: DG.Column[] = [seq1Col, seq2Col, diffCol, act1Col, act2Col, deltaCol];
+
+    // Add extra columns (two per extra column: one for seq1, one for seq2)
+    for (let e = 0; e < extraColumns.length; e++) {
+      const eName = extraColumns[e].name;
+      const eType = extraColumns[e].type;
+      columns.push(DG.Column.fromList(eType as DG.COLUMN_TYPE, `Seq 1 ${eName}`, extraData1[e]));
+      columns.push(DG.Column.fromList(eType as DG.COLUMN_TYPE, `Seq 2 ${eName}`, extraData2[e]));
+    }
+
+    const df = DG.DataFrame.fromColumns(columns);
+    df.name = 'Mutation Cliffs';
+    grok.shell.addTableView(df);
+  }
+
+  /** Exports the invariant map as a new table view: monomer × position counts. */
+  exportInvariantMap(): void {
+    const stats = this.monomerPositionStats;
+    const uniqueMonomers = new Set<string>();
+    const positionNames: string[] = [];
+    for (const pos of Object.keys(stats)) {
+      if (pos === 'general')
+        continue;
+      positionNames.push(pos);
+      const posStats = stats[pos]!;
+      for (const monomer of Object.keys(posStats)) {
+        if (monomer === 'general')
+          continue;
+        uniqueMonomers.add(monomer);
+      }
+    }
+
+    const monomersArray = Array.from(uniqueMonomers).sort();
+    const monomerCol = DG.Column.fromStrings(C.COLUMNS_NAMES.MONOMER, monomersArray);
+    const columns: DG.Column[] = [monomerCol];
+
+    for (const pos of positionNames) {
+      const posStats = stats[pos]!;
+      const counts = monomersArray.map((m) => posStats[m]?.count ?? 0);
+      columns.push(DG.Column.fromList(DG.COLUMN_TYPE.INT, pos, counts));
+    }
+
+    const df = DG.DataFrame.fromColumns(columns);
+    df.name = 'Invariant Map';
+    grok.shell.addTableView(df);
+  }
+
   /** Removes all the active subscriptions. */
   detach(): void {
     this.subs.forEach((sub) => sub.unsubscribe());
@@ -567,6 +726,9 @@ export abstract class SARViewer extends DG.JsViewer implements ISARViewer {
         if (!a || !a.causedBy || !a.args || !a.args.menu || !a.causedBy.target || !(a.causedBy.target instanceof HTMLElement) || !this.root.contains(a.causedBy.target))
           return;
         const menu = a.args.menu as DG.Menu;
+        const exportGroup = menu.group('Export');
+        exportGroup.item('Export Mutation Cliffs...', () => this.exportMutationCliffs());
+        exportGroup.item('Export Invariant Map', () => this.exportInvariantMap());
         getMonomerLibHelper().then((lh) => {
           const lib = lh.getMonomerLib();
           const mSymbols = lib.getMonomerSymbolsByType(PolymerTypes.PEPTIDE);
@@ -748,6 +910,8 @@ export class MonomerPosition extends SARViewer {
    */
   onPropertyChanged(property: DG.Property): void {
     super.onPropertyChanged(property);
+    if (!this.dataFrame)
+      return;
     switch (property.name) {
     case SAR_PROPERTIES.SEQUENCE:
       this._invariantMapSelection = null;
@@ -915,6 +1079,7 @@ export class MonomerPosition extends SARViewer {
       sumGridCol.visible = false;
     const monomerCol = monomerPositionDf.getCol(C.COLUMNS_NAMES.MONOMER);
     CR.setMonomerRenderer(monomerCol, this.alphabet, true);
+    this.propagateCanonicalizerTag(monomerCol);
     this.cacheInvariantMapColors();
 
     grid.onCellRender.subscribe((args: DG.GridCellRenderArgs) => renderCell(args, this,
@@ -1115,7 +1280,7 @@ export class MonomerPosition extends SARViewer {
   /** Renders the MonomerPosition viewer body. */
   render(): void {
     $(this.root).empty();
-    if (!this.activityColumnName || !this.sequenceColumnName) {
+    if (!this.dataFrame || !this.activityColumnName || !this.sequenceColumnName) {
       this.root.appendChild(ui.divText('Please, select a sequence and activity columns in the viewer properties'));
       return;
     }
@@ -1342,6 +1507,7 @@ export class MostPotentResidues extends SARViewer {
 
     // Setting Monomer column renderer
     CR.setMonomerRenderer(monomerCol, this.alphabet, true);
+    this.propagateCanonicalizerTag(monomerCol);
     grid.onCellRender.subscribe(
       (args: DG.GridCellRenderArgs) => renderCell(args, this, false, undefined, undefined));
 
@@ -1474,7 +1640,7 @@ export class MostPotentResidues extends SARViewer {
   /** Renders the MostPotentResidues viewer body. */
   render(): void {
     $(this.root).empty();
-    if (!this.activityColumnName || !this.sequenceColumnName) {
+    if (!this.dataFrame || !this.activityColumnName || !this.sequenceColumnName) {
       this.root.appendChild(ui.divText('Please, select a sequence and activity columns in the viewer properties'));
       return;
     }

package/src/widgets/manual-alignment.ts

@@ -24,7 +24,7 @@ export function manualAlignmentWidget(alignedSequenceCol: DG.Column<string>, cur
     const affectedRowIndex = currentDf.currentRowIdx;
     alignedSequenceCol.set(affectedRowIndex, newSequence);
     for (let i = 0; i < splitSequence.length; i++) {
-      const part = splitSequence.getCanonical(i);
+      const part = splitSequence.getOriginal(i);
       if (currentDf.col(i.toString()) !== null)
         currentDf.set(i.toString(), affectedRowIndex, part);
     }