@datagrok/peptides 0.8.9 → 0.8.13

package/src/utils/peptide-similarity-space.ts

@@ -10,21 +10,7 @@ import {
 } from '@datagrok-libraries/ml/src/workers/dimensionality-reducing-worker-creator';
 import {Measure, StringMetrics} from '@datagrok-libraries/ml/src/typed-metrics';
 import {Coordinates} from '@datagrok-libraries/utils/src/type-declarations';
-
-/**
- * Finds a column with an activity.
- *
- * @param {DG.DataFrame} table The data frame to search for.
- * @return {(string | null)} Column name or null if not found.
- */
-function inferActivityColumnsName(table: DG.DataFrame): string | null {
-  const re = /activity|ic50/i;
-  for (const name of table.columns.names()) {
-    if (name.match(re))
-      return name;
-  }
-  return null;
-}
+import * as C from './constants';
 
 /**
  * Cast an aligned sequences column to clean sequences.
@@ -52,14 +38,13 @@ export function cleanAlignedSequencesColumn(col: DG.Column): Array<string> {
  * @return {Promise<DG.ScatterPlotViewer>} A viewer.
  */
 export async function createPeptideSimilaritySpaceViewer(
-  table: DG.DataFrame, alignedSequencesColumn: DG.Column, method: string, measure: string, cyclesCount: number,
-  view: DG.TableView | null, activityColumnName?: string | null): Promise<DG.ScatterPlotViewer> {
+  table: DG.DataFrame, method: string, measure: string, cyclesCount: number, view?: DG.TableView, col?: DG.Column,
+): Promise<DG.ScatterPlotViewer> {
   const pi = DG.TaskBarProgressIndicator.create('Creating embedding...');
 
-  activityColumnName = activityColumnName ?? inferActivityColumnsName(table);
-
   const axesNames = ['~X', '~Y', '~MW'];
-  const columnData = alignedSequencesColumn.toList().map((v, _) => AlignedSequenceEncoder.clean(v));
+  const columnData = (col ?? table.getCol(C.COLUMNS_NAMES.ALIGNED_SEQUENCE)).toList()
+    .map((v) => AlignedSequenceEncoder.clean(v));
 
   const embcols = await createDimensinalityReducingWorker(
     {data: columnData, metric: measure as StringMetrics}, method, cyclesCount);
@@ -94,11 +79,13 @@ export async function createPeptideSimilaritySpaceViewer(
   }
 
   const viewerOptions = {
-    x: '~X', y: '~Y', color: activityColumnName ?? '~MW', size: '~MW', title: 'Peptide Space', showYSelector: false,
-    showXSelector: false, showColorSelector: false, showSizeSelector: false,
+    x: '~X', y: '~Y', color: C.COLUMNS_NAMES.ACTIVITY ?? '~MW', size: '~MW', title: 'Peptide Space',
+    showYSelector: false, showXSelector: false, showColorSelector: false, showSizeSelector: false,
   };
   const viewer = table.plot.scatter(viewerOptions);
 
+  view?.dockManager.dock(viewer, DG.DOCK_TYPE.RIGHT, null, 'Peptide Space viewer');
+
   pi.close();
   return viewer;
 }
@@ -131,7 +118,8 @@ export class PeptideSimilaritySpaceWidget {
     this.availableMetrics = Measure.getMetricByDataType('String');
     this.method = this.availableMethods[0];
     this.metrics = this.availableMetrics[0];
-    this.currentDf = alignedSequencesColumn.dataFrame;
+    const df = alignedSequencesColumn.dataFrame;
+    this.currentDf = df.clone(df.filter);
     this.alignedSequencesColumn = alignedSequencesColumn;
     this.viewer = ui.div([]);
     this.view = view;
@@ -145,14 +133,7 @@ export class PeptideSimilaritySpaceWidget {
    */
   public async drawViewer(): Promise<DG.Viewer> {
     const viewer = await createPeptideSimilaritySpaceViewer(
-      this.currentDf,
-      this.alignedSequencesColumn,
-      this.method,
-      this.metrics,
-      this.cycles,
-      null,
-      null,
-    );
+      this.currentDf, this.method, this.metrics, this.cycles, undefined, this.alignedSequencesColumn);
     viewer.root.style.width = 'auto';
     return viewer;
   }

package/src/utils/types.ts

@@ -0,0 +1,10 @@
+import * as DG from 'datagrok-api/dg';
+
+export type SubstitutionCases = {[aar: string]: number[][][]};
+export type SubstitutionTooltips = { [aar: string]: {}[][]; };
+export type DataFrameDict = {[key: string]: DG.DataFrame};
+
+export namespace BarChart {
+  export type BarPart = {colName : string, aaName : string};
+  export type BarStatsObject = {name: string, count: number, selectedCount: number};
+}

package/src/viewers/logo-viewer.ts

@@ -7,6 +7,7 @@ import * as logojs from 'logojs-react';
 // import {splitAlignedPeptides} from '../utils/split-aligned';
 import {ChemPalette} from '../utils/chem-palette';
 import {PeptidesController} from '../peptides';
+import * as C from '../utils/constants';
 
 /**
  * Logo viewer.
@@ -34,7 +35,7 @@ export class Logo extends DG.JsViewer {
   constructor() {
     super();
     this.initialized = false;
-    this.colSemType = this.string('colSemType', 'alignedSequence');
+    this.colSemType = this.string('colSemType', C.SEM_TYPES.ALIGNED_SEQUENCE);
 
     this.splitted = null;
     this.ppm = [];

package/src/viewers/peptide-space-viewer.ts

@@ -0,0 +1,121 @@
+import * as ui from 'datagrok-api/ui';
+import * as grok from 'datagrok-api/grok';
+import * as DG from 'datagrok-api/dg';
+
+import $ from 'cash-dom';
+
+import {getSequenceMolecularWeight} from '../utils/molecular-measure';
+import {AlignedSequenceEncoder} from '@datagrok-libraries/bio/src/sequence-encoder';
+import {
+  createDimensinalityReducingWorker,
+} from '@datagrok-libraries/ml/src/workers/dimensionality-reducing-worker-creator';
+import {StringMetrics} from '@datagrok-libraries/ml/src/typed-metrics';
+import {Coordinates} from '@datagrok-libraries/utils/src/type-declarations';
+import * as C from '../utils/constants';
+import {PeptidesController} from '../peptides';
+
+export class PeptideSpaceViewer extends DG.JsViewer {
+  method: string;
+  measure: string;
+  cyclesCount: number;
+  // controller: PeptidesController | null = null;
+  customProperties = new Set(['method', 'measure', 'cyclesCount']);
+  isEmbeddingCreating: boolean = false;
+
+  constructor() {
+    super();
+
+    const methodChoices = ['UMAP', 't-SNE', 'SPE', 'pSPE', 'OriginalSPE'];
+    this.method = this.addProperty('method', DG.TYPE.STRING, 'UMAP', {choices: methodChoices});
+    const measureChoices = ['Levenshtein', 'Jaro-Winkler'];
+    this.measure = this.addProperty('measure', DG.TYPE.STRING, 'Levenshtein', {choices: measureChoices});
+    this.cyclesCount = this.addProperty('cyclesCount', DG.TYPE.INT, 100);
+  }
+
+  async onFrameAttached(dataFrame: DG.DataFrame) {
+    super.onFrameAttached(dataFrame);
+
+    // this.controller = await PeptidesController.getInstance(this.dataFrame!);
+
+    await this.render(this.dataFrame!.temp[C.EMBEDDING_STATUS]);
+  }
+
+  async onPropertyChanged(property: DG.Property | null) {
+    super.onPropertyChanged(property);
+
+
+    await this.render(this.customProperties.has(property?.name ?? '') || this.dataFrame!.temp[C.EMBEDDING_STATUS]);
+  }
+
+  async render(computeData=false) {
+    if (computeData && !this.isEmbeddingCreating) {
+      this.isEmbeddingCreating = true;
+      $(this.root).empty();
+      const viewerHost = ui.waitBox(async () => {
+        await computeWeights(this.dataFrame!, this.method, this.measure, this.cyclesCount);
+
+        const viewerOptions = {
+          x: '~X', y: '~Y', color: C.COLUMNS_NAMES.ACTIVITY_SCALED ?? '~MW', size: '~MW', title: 'Peptide Space',
+          showYSelector: false, showXSelector: false, showColorSelector: false, showSizeSelector: false,
+        };
+        const viewerRoot = this.dataFrame!.plot.scatter(viewerOptions).root;
+        viewerRoot.style.width = 'auto';
+        this.isEmbeddingCreating = false;
+        viewerHost.style.paddingLeft = 'unset';
+        return viewerRoot;
+      }) as HTMLDivElement;
+      viewerHost.style.paddingLeft = '45%';
+      this.root.appendChild(viewerHost);
+    }
+  }
+}
+
+export async function computeWeights(
+  table: DG.DataFrame, method: string, measure: string, cyclesCount: number, col?: DG.Column,
+): Promise<void> {
+  const pi = DG.TaskBarProgressIndicator.create('Creating embedding...');
+  try {
+    const axesNames = ['~X', '~Y', '~MW'];
+    const columnData = (col ?? table.getCol(C.COLUMNS_NAMES.ALIGNED_SEQUENCE)).toList()
+      .map((v) => AlignedSequenceEncoder.clean(v));
+
+    const embcols = await createDimensinalityReducingWorker(
+      {data: columnData, metric: measure as StringMetrics}, method, cyclesCount);
+
+    const columns = Array.from(
+      embcols as Coordinates, (v: Float32Array, k) => DG.Column.fromFloat32Array(axesNames[k], v));
+
+    function _getMW(sequences: string[]) {
+      const mw: Float32Array = new Float32Array(sequences.length);
+
+      mw.map((_, index) => getSequenceMolecularWeight(sequences[index] ?? ''));
+
+      return mw;
+    }
+
+    columns.push(DG.Column.fromFloat32Array('~MW', _getMW(columnData)));
+
+    const edf = DG.DataFrame.fromColumns(columns);
+
+    // Add new axes.
+    for (const axis of axesNames) {
+      const col = table.col(axis);
+      const newCol = edf.getCol(axis);
+
+      // if (col != null) {
+      //   for (let i = 0; i < newCol.length; ++i) {
+      //     const v = newCol.get(i);
+      //     table.set(axis, i, v);
+      //   }
+      // } else
+      //   table.columns.insert(newCol);
+      const columnList = table.columns as DG.ColumnList;
+      col !== null ? columnList.replace(col, newCol) : columnList.insert(newCol);
+    }
+  } catch (error) {
+    grok.shell.error('Could not compute embeddings. See console for details.');
+    console.error(error);
+  } finally {
+    pi.close();
+  }
+}