@datagrok/peptides 1.27.5 → 1.27.7

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@datagrok/peptides",
   "friendlyName": "Peptides",
-  "version": "1.27.5",
+  "version": "1.27.7",
   "author": {
     "name": "Davit Rizhinashvili",
     "email": "drizhinashvili@datagrok.ai"
@@ -13,7 +13,7 @@
     "directory": "packages/Peptides"
   },
   "dependencies": {
-    "@datagrok-libraries/bio": "^5.61.3",
+    "@datagrok-libraries/bio": "^5.63.2",
     "@datagrok-libraries/math": "^1.2.6",
     "@datagrok-libraries/ml": "^6.10.7",
     "@datagrok-libraries/statistics": "^1.2.12",
@@ -21,7 +21,7 @@
     "@datagrok-libraries/utils": "^4.6.9",
     "@webgpu/types": "^0.1.40",
     "cash-dom": "^8.1.5",
-    "datagrok-api": "^1.25.0",
+    "datagrok-api": "^1.26.0",
     "file-loader": "^6.2.0",
     "rxjs": "^6.5.5",
     "uuid": "^10.0.0",
@@ -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/package.g.ts

@@ -14,8 +14,8 @@ export function Peptides() : any {
 
 //name: Bio Peptides
 //top-menu: Bio | Analyze | SAR...
-export function peptidesDialog() : any {
-  return PackageFunctions.peptidesDialog();
+export async function peptidesDialog() : Promise<any> {
+  return await PackageFunctions.peptidesDialog();
 }
 
 //input: viewer v 
@@ -28,8 +28,8 @@ export async function testInitFunctionPeptides(v: any) : Promise<void> {
 //input: column col { semType: Macromolecule }
 //output: widget result
 //meta.role: widgets,panel
-export function peptidesPanel(col: DG.Column) : any {
-  return PackageFunctions.peptidesPanel(col);
+export async function peptidesPanel(col: DG.Column) : Promise<any> {
+  return await PackageFunctions.peptidesPanel(col);
 }
 
 //name: Sequence Variability Map

package/src/package.ts

@@ -128,7 +128,7 @@ export class PackageFunctions {
     'name': 'Bio Peptides',
     'outputs': [],
   })
-  static peptidesDialog(): DG.Dialog | null {
+  static async peptidesDialog(): Promise<DG.Dialog | null> {
     if (!grok.shell.t || !grok.shell.t.columns.bySemType('Macromolecule')?.length) {
       grok.shell.warning('SAR Analysis requires an active table with Macromolecule column');
       return null;
@@ -139,7 +139,7 @@ export class PackageFunctions {
       return null;
     }
 
-    const analyzeObject = analyzePeptidesUI(grok.shell.t);
+    const analyzeObject = await analyzePeptidesUI(grok.shell.t);
     const dialog = ui.dialog('Analyze Peptides').add(analyzeObject.host).onOK(async () => {
       const startSuccess = analyzeObject.callback();
       if (!startSuccess)
@@ -163,11 +163,11 @@ export class PackageFunctions {
     meta: {role: 'widgets'},
     tags: ['widgets', 'panel'],
   })
-  static peptidesPanel(
-    @grok.decorators.param({'options': {'semType': 'Macromolecule'}}) col: DG.Column): DG.Widget {
+  static async peptidesPanel(
+    @grok.decorators.param({'options': {'semType': 'Macromolecule'}}) col: DG.Column): Promise<DG.Widget> {
     if (!col.dataFrame || !DG.Utils.firstOrNull(col.dataFrame.columns.numerical))
       return new DG.Widget(ui.divText('SAR Analysis requires an active table with at least one numerical column for activity'));
-    const analyzeObject = analyzePeptidesUI(col.dataFrame, col);
+    const analyzeObject = await analyzePeptidesUI(col.dataFrame, col);
     return new DG.Widget(analyzeObject.host);
   }
 

package/src/viewers/sar-viewer.ts

@@ -42,6 +42,8 @@ import {getMonomerLibHelper} from '@datagrok-libraries/bio/src/types/monomer-lib
 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',
@@ -543,6 +545,152 @@ 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 +715,9 @@ 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);

package/src/widgets/peptides.ts

@@ -12,6 +12,7 @@ import {ALIGNMENT, NOTATION, TAGS as bioTAGS} from '@datagrok-libraries/bio/src/
 import {ILogoSummaryTable, LogoSummaryTable} from '../viewers/logo-summary';
 import {MmDistanceFunctionsNames} from '@datagrok-libraries/ml/src/macromolecule-distance-functions';
 import {MCL_INPUTS} from './settings';
+import {createSequenceColumnInput, ISequenceColumnInput} from '@datagrok-libraries/bio/src/utils/sequence-column-input';
 //@ts-ignore
 import '../styles.css';
 export type DialogParameters = { host: HTMLElement, callback: () => Promise<boolean> };
@@ -22,10 +23,10 @@ export type DialogParameters = { host: HTMLElement, callback: () => Promise<bool
  * @param [col] - Peptides column
  * @return - UI host and analysis start callback
  */
-export function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): DialogParameters {
+export async function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): Promise<DialogParameters> {
   const mclOptions = new type.MCLSettings();
   const logoHost = ui.div();
-  let seqColInput: DG.InputBase | null = null;
+  let sequenceColInput: ISequenceColumnInput | null = null;
   if (typeof col === 'undefined') {
     // Building UI for starting analysis from dialog (top menu)
     const potentialCol = DG.Utils.firstOrNull(
@@ -36,7 +37,7 @@ export function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): Di
       grok.shell.info('Sequences column contains missing values. They will be ignored during analysis');
 
 
-    seqColInput = ui.input.column('Sequence', {table: df, value: potentialCol, onValueChanged: (value) => {
+    sequenceColInput = await createSequenceColumnInput('Sequence', {table: df, value: potentialCol, onValueChanged: (value) => {
       $(logoHost).empty().append(ui.wait(async () => {
         const viewer = await df.plot.fromType('WebLogo', {sequenceColumnName: value.name});
         viewer.root.style.setProperty('height', '130px');
@@ -45,7 +46,7 @@ export function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): Di
       if (value.stats.missingValueCount !== 0)
         grok.shell.info('Sequences column contains missing values. They will be ignored during analysis');
     }, filter: (col: DG.Column) => col.semType === DG.SEMTYPE.MACROMOLECULE, nullable: false});
-    seqColInput.setTooltip('Macromolecule column in FASTA, HELM or separated format');
+    sequenceColInput.inputBase.setTooltip('Macromolecule column in FASTA, HELM or separated format');
   }
 
   let funcs = DG.Func.find({package: 'Bio', name: 'webLogoViewer'});
@@ -117,8 +118,8 @@ export function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): Di
 
 
   const inputsList = [activityColumnChoice, activityScalingMethod, clustersColumnChoice, generateClustersInput];
-  if (seqColInput !== null)
-    inputsList.splice(0, 0, seqColInput);
+  if (sequenceColInput !== null)
+    inputsList.splice(0, 0, sequenceColInput.inputBase);
 
   // ### MCL INPUTS ###
   const similarityThresholdInput = ui.input.float(MCL_INPUTS.THRESHOLD, {
@@ -188,7 +189,7 @@ export function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): Di
   const bitsetChanged = df.filter.onChanged.subscribe(() => activityScalingMethodState());
 
   const startAnalysisCallback = async (): Promise<boolean> => {
-    const sequencesCol = col ?? seqColInput!.value;
+    const sequencesCol = col ?? sequenceColInput!.value;
     bitsetChanged.unsubscribe();
     if (sequencesCol) {
       const model = await startAnalysis(activityColumnChoice.value!, sequencesCol, clustersColumnChoice.value, df,
@@ -210,7 +211,7 @@ export function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): Di
   }
 
   $(logoHost).empty().append(ui.wait(async () => {
-    const viewer = await df.plot.fromType('WebLogo', {sequenceColumnName: col?.name ?? seqColInput!.value!.name});
+    const viewer = await df.plot.fromType('WebLogo', {sequenceColumnName: col?.name ?? sequenceColInput!.value!.name});
     viewer.root.style.setProperty('height', '130px');
     return viewer.root;
   }));
@@ -218,9 +219,9 @@ export function analyzePeptidesUI(df: DG.DataFrame, col?: DG.Column<string>): Di
   const mainHost = ui.divV([
     logoHost,
     ui.splitH([
-      ui.splitV(inputElements),
+      ui.splitV(inputElements, {style: {flex: '0 1 auto'}}),
       histogramHost,
-    ], {style: {height: bottomHeight, minWidth: '500px', maxWidth: '600px'}}),
+    ], {style: {height: bottomHeight, minWidth: '600px', maxWidth: '750px'}}),
     mclInputsHost,
   ]);
   return {host: mainHost, callback: startAnalysisCallback};