@datagrok/peptides 0.3.0 → 0.5.6

package/.eslintrc.json

@@ -0,0 +1,29 @@
+{
+  "env": {
+    "browser": true,
+    "es2021": true
+  },
+  "extends": [
+    "google"
+  ],
+  "parser": "@typescript-eslint/parser",
+  "parserOptions": {
+    "ecmaVersion": 12,
+    "sourceType": "module"
+  },
+  "plugins": [
+    "@typescript-eslint"
+  ],
+  "rules": {
+    "indent": [
+      "error",
+      2
+    ],
+    "max-len": [
+      "error",
+      120
+    ],
+    "spaced-comment": "off",
+    "require-jsdoc": "off"
+  }
+}

package/package.json

@@ -1,20 +1,20 @@
 {
 	"name": "@datagrok/peptides",
-	"version": "0.3.0",
+	"version": "0.5.6",
 	"description": "",
 	"dependencies": {
 		"@keckelt/tsne": "^1.0.2",
 		"cash-dom": "latest",
 		"d3": "latest",
-		"datagrok-api": ">0.95.4",
+		"datagrok-api": ">=0.104.0",
 		"dayjs": "latest",
 		"jaro-winkler-typescript": "^1.0.1",
 		"jstat": "^1.9.5",
 		"logojs-react": "^2.1.1",
 		"rxjs": "^6.5.5",
 		"umap-js": "^1.3.3",
-		"@datagrok-libraries/utils": ">=0.0.10",
-		"@datagrok-libraries/statistics": ">=0.1.4",
+		"@datagrok-libraries/utils": ">=0.0.13",
+		"@datagrok-libraries/statistics": ">=0.1.5",
 		"@types/d3": "^7.0.0",
 		"@types/jquery": "^3.5.6"
 	},

package/src/describe.ts

@@ -1,15 +1,59 @@
-// eslint-disable-next-line no-unused-vars
-import * as grok from 'datagrok-api/grok';
 import * as ui from 'datagrok-api/ui';
 import * as DG from 'datagrok-api/dg';
 import {splitAlignedPeptides} from './utils/split-aligned';
 import {tTest} from '@datagrok-libraries/statistics/src/tests';
-import {fdrcorrection} from '@datagrok-libraries/statistics/src/multiple-tests.js';
+import {fdrcorrection} from '@datagrok-libraries/statistics/src/multiple-tests';
 import {ChemPalette} from './utils/chem-palette';
 import {setAARRenderer} from './utils/cell-renderer';
 
 const cp = new ChemPalette('grok');
 
+const aarGroups = {
+  'R': 'PC',
+  'H': 'PC',
+  'K': 'PC',
+  'D': 'NC',
+  'E': 'NC',
+  'S': 'U',
+  'T': 'U',
+  'N': 'U',
+  'Q': 'U',
+  'C': 'SC',
+  'U': 'SC',
+  'G': 'SC',
+  'P': 'SC',
+  'A': 'H',
+  'V': 'H',
+  'I': 'H',
+  'L': 'H',
+  'M': 'H',
+  'F': 'H',
+  'Y': 'H',
+  'W': 'H',
+  '-': '-',
+};
+
+const groupDescription: {[key: string]: {'description': string, 'aminoAcids': string[]}} = {
+  'PC': {'description': 'Positive Amino Acids, with Electrically Charged Side Chains', 'aminoAcids': ['R', 'H', 'K']},
+  'NC': {'description': 'Negative Amino Acids, with Electrically Charged Side Chains', 'aminoAcids': ['D', 'E']},
+  'U': {'description': 'Amino Acids with Polar Uncharged Side Chains', 'aminoAcids': ['S', 'T', 'N', 'Q']},
+  'SC': {'description': 'Special Cases', 'aminoAcids': ['C', 'U', 'G', 'P']},
+  'H': {
+    'description': 'Amino Acids with Hydrophobic Side Chain',
+    'aminoAcids': ['A', 'V', 'I', 'L', 'M', 'F', 'Y', 'W'],
+  },
+  '-': {'description': 'Unknown Amino Acid', 'aminoAcids': ['-']},
+};
+
+/*function customGridColumnHeader(cell: DG.GridCell) {
+  if (cell.isColHeader && cell.tableColumn != null) {
+    if (highlightedColumns.includes(parseInt(cell.tableColumn.name))) {
+      cell.style.backColor = 0xff1f77b4;
+    }
+  }
+}*/
+
+//TODO: decomposition!
 export async function describe(
   df: DG.DataFrame,
   activityColumn: string,
@@ -17,13 +61,15 @@ export async function describe(
   sourceGrid: DG.Grid,
   twoColorMode: boolean,
   initialBitset: DG.BitSet | null,
-): Promise<[DG.Grid, DG.Grid, DG.DataFrame]> {
+  grouping: boolean,
+): Promise<[DG.Grid, DG.Grid, DG.DataFrame, {[key: string]: string}]> {
   //Split the aligned sequence into separate AARs
   let splitSeqDf: DG.DataFrame | undefined;
   let invalidIndexes: number[];
   const col: DG.Column = df.columns.bySemType('alignedSequence');
   [splitSeqDf, invalidIndexes] = splitAlignedPeptides(col);
   splitSeqDf.name = 'Split sequence';
+
   const positionColumns = splitSeqDf.columns.names();
   const activityColumnScaled = `${activityColumn}Scaled`;
   const renderColNames: string[] = splitSeqDf.columns.names();
@@ -50,6 +96,7 @@ export async function describe(
       setAARRenderer(col, sourceGrid);
     }
   }
+
   if (sourceGrid) {
     const colNames:string[] = [];
     for (let i = 0; i < sourceGrid.columns.length; i++) {
@@ -105,6 +152,17 @@ export async function describe(
 
   let matrixDf = splitSeqDf.unpivot([activityColumnScaled], positionColumns, positionColName, aminoAcidResidue);
 
+  //TODO: move to chem palette
+  let groupMapping: {[key: string]: string} = {};
+  if (grouping) {
+    groupMapping = aarGroups;
+    const aarCol = matrixDf.getCol(aminoAcidResidue);
+    aarCol.init((index) => groupMapping[aarCol.get(index)[0]] ?? '-');
+    aarCol.compact();
+  } else {
+    Object.keys(aarGroups).forEach((value) => groupMapping[value] = value);
+  }
+
   //statistics for specific AAR at a specific position
   matrixDf = matrixDf.groupBy([positionColName, aminoAcidResidue])
     .add('count', activityColumnScaled, 'Count')
@@ -135,14 +193,14 @@ export async function describe(
     AAR = matrixDf.get(aminoAcidResidue, i);
 
     //@ts-ignore
-    splitSeqDf.rows.select((row) => row[position] === AAR);
+    splitSeqDf.rows.select((row) => groupMapping[row[position]] === AAR);
     currentActivity = splitSeqDf
       .clone(splitSeqDf.selection, [activityColumnScaled])
       .getCol(activityColumnScaled)
       .toList();
 
     //@ts-ignore
-    splitSeqDf.rows.select((row) => row[position] !== AAR);
+    splitSeqDf.rows.select((row) => groupMapping[row[position]] !== AAR);
     otherActivity = splitSeqDf
       .clone(splitSeqDf.selection, [activityColumnScaled])
       .getCol(activityColumnScaled)
@@ -190,11 +248,9 @@ export async function describe(
   aarList.sort((first, second) => getWeight(second) - getWeight(first));
 
   matrixDf.getCol(aminoAcidResidue).setCategoryOrder(aarList);
-  //const sequenceDf = segregateBestAtAllCateg(statsDf, twoColorMode);
 
   // SAR vertical table (naive, choose best Mean difference from pVals <= 0.01)
   // TODO: aquire ALL of the positions
-
   let sequenceDf = statsDf.groupBy(['Mean difference', aminoAcidResidue, positionColName, 'Count', 'Ratio', 'pValue'])
     .where('pValue <= 0.1')
     .aggregate();
@@ -224,17 +280,14 @@ export async function describe(
   SARVgrid.col('pValue')!.format = 'four digits after comma';
   SARVgrid.col('pValue')!.name = 'P-Value';
 
-  //FIXME: looks inefficient
-  for (const col of matrixDf.columns) {
-    if (col.name === aminoAcidResidue) {
-      setAARRenderer(col, SARgrid);
-      break;
+  if (!grouping) {
+    let tempCol = matrixDf.columns.byName(aminoAcidResidue);
+    if (tempCol) {
+      setAARRenderer(tempCol, SARgrid);
     }
-  }
-  for (const col of sequenceDf.columns) {
-    if (col.name === aminoAcidResidue) {
-      setAARRenderer(col, SARVgrid);
-      break;
+    tempCol = sequenceDf.columns.byName(aminoAcidResidue);
+    if (tempCol) {
+      setAARRenderer(tempCol, SARgrid);
     }
   }
 
@@ -251,19 +304,6 @@ export async function describe(
       return;
     }
 
-    // if (args.cell.isColHeader) {
-    //   if (args.cell.gridColumn.name != aminoAcidResidue) {
-    //     const textSize = args.g.measureText(args.cell.gridColumn.name);
-    //     args.g.fillStyle = '#4b4b4a';
-    //     args.g.fillText(
-    //       args.cell.gridColumn.name,
-    //       args.bounds.x + (args.bounds.width - textSize.width) / 2,
-    //       args.bounds.y + (textSize.actualBoundingBoxAscent + textSize.actualBoundingBoxDescent),
-    //     );
-    //   }
-    //   args.preventDefault();
-    // }
-
     if (
       args.cell.isTableCell &&
       args.cell.tableRowIndex !== null &&
@@ -357,28 +397,39 @@ export async function describe(
     }
     if (
       !cell.isColHeader &&
-        cell.tableColumn !== null &&
-        cell.tableColumn.name == aminoAcidResidue &&
-        cell.cell.value !== null &&
-        cell.tableRowIndex !== null
+      cell.tableColumn !== null &&
+      cell.tableColumn.name == aminoAcidResidue &&
+      cell.cell.value !== null &&
+      cell.tableRowIndex !== null
     ) {
-      cp.showTooltip(cell, x, y);
+      if (grouping) {
+        const currentGroup = groupDescription[cell.cell.value];
+        const divText = ui.divText('Amino Acids in this group: ' + currentGroup['aminoAcids'].join(', '));
+        ui.tooltip.show(ui.divV([ui.h3(currentGroup['description']), divText]), x, y);
+      } else {
+        cp.showTooltip(cell, x, y);
+      }
     }
     return true;
   };
   SARgrid.onCellTooltip(onCellTooltipFunc);
   SARVgrid.onCellTooltip(onCellTooltipFunc);
 
-  sourceGrid.onCellPrepare((cell) => {
+  sourceGrid.onCellPrepare((cell: DG.GridCell) => {
     const currentRowIndex = cell.tableRowIndex;
     if (currentRowIndex && invalidIndexes.includes(currentRowIndex) && !cell.isRowHeader) {
       cell.style.backColor = DG.Color.lightLightGray;
     }
   });
 
-  // for (const col of matrixDf.columns.names()) {
-  //   SARgrid.col(col)!.width = SARgrid.props.rowHeight;
-  // }
+  for (const col of matrixDf.columns.names()) {
+    SARgrid.col(col)!.width = SARgrid.props.rowHeight;
+  }
+
+  if (grouping) {
+    SARgrid.col(aminoAcidResidue)!.name = 'Groups';
+    SARVgrid.col(aminoAcidResidue)!.name = 'Groups';
+  }
 
-  return [SARgrid, SARVgrid, statsDf];
+  return [SARgrid, SARVgrid, statsDf, groupMapping];
 }

package/src/package.ts

@@ -15,7 +15,7 @@ import {PeptideSimilaritySpaceWidget} from './utils/peptide-similarity-space';
 import {manualAlignmentWidget} from './widgets/manual-alignment';
 import {SARViewer, SARViewerVertical} from './viewers/sar-viewer';
 import {peptideMoleculeWidget} from './widgets/peptide-molecule';
-import {correlationAnalysisPlots} from './utils/correlation-analysis';
+import {SpiralPlot} from './viewers/spiral-plot';
 
 export const _package = new DG.Package();
 let tableGrid: DG.Grid;
@@ -25,35 +25,13 @@ let view: DG.TableView;
 
 async function main(chosenFile: string) {
   const pi = DG.TaskBarProgressIndicator.create('Loading Peptides');
-  //let peptides =
-  //  await grok.data.loadTable('https://datagrok.jnj.com/p/ejaeger.il23peptideidp5562/il-23_peptide_idp-5562');
   const path = _package.webRoot + 'files/' + chosenFile;
   const peptides = (await grok.data.loadTable(path));
   peptides.name = 'Peptides';
   peptides.setTag('dataType', 'peptides');
   const view = grok.shell.addTableView(peptides);
   tableGrid = view.grid;
-  // peptides.onSemanticTypeDetecting.subscribe((_: any) => {
-  //   const regexp = new RegExp(/^([^-^\n]*-){2,49}(\w|\(|\))+$/);
-  //   for (const col of peptides.columns) {
-  //     col.semType = DG.Detector.sampleCategories(col, (s: any) => regexp.test(s.trim())) ? 'alignedSequence' : null;
-  //     if (col.semType == 'alignedSequence') {
-  //       expandColumn(col, tableGrid, (ent)=>{
-  //         const subParts:string[] = ent.split('-');
-  //         // eslint-disable-next-line no-unused-vars
-  //         const [text, _] = processSequence(subParts);
-  //         let textSize = 0;
-  //         text.forEach((aar)=>{
-  //           textSize += aar.length;
-  //         });
-  //         return textSize;
-  //       });
-  //     }
-  //   }
-  // });
-
   view.name = 'PeptidesView';
-
   grok.shell.windows.showProperties = true;
 
   pi.close();
@@ -67,7 +45,6 @@ export function Peptides() {
 
   const appDescription = ui.info(
     [
-      // ui.divText('\n To start the application :', {style: {'font-weight': 'bolder'}}),
       ui.list([
         '- automatic recognition of peptide sequences',
         '- native integration with tons of Datagrok out-of-the box features (visualization, filtering, clustering, ' +
@@ -197,18 +174,19 @@ export function manualAlignment(monomer: string) {
 //input: column col {semType: alignedSequence}
 //output: widget result
 export async function peptideSpacePanel(col: DG.Column): Promise<DG.Widget> {
-  const widget = new PeptideSimilaritySpaceWidget(col);
+  const widget = new PeptideSimilaritySpaceWidget(col, view ?? grok.shell.v);
   return await widget.draw();
 }
 
-//name: Correllation analysis
-export async function correlationAnalysis() {
-  view = (grok.shell.v as DG.TableView);
-
-  const df = await grok.data.files.openTable('Demo:TestJobs:Files:DemoFiles/bio/peptides.csv');
-  const tview = grok.shell.addTableView(df);
-  const [cpviewer, bpviewer] = correlationAnalysisPlots(df.getCol('AlignedSequence'));
-
-  tview.dockManager.dock(cpviewer, 'right');
-  tview.dockManager.dock(bpviewer, 'down');
+//name: Spiral Plot
+////input: dataframe table
+////input: column activity
+//tags: viewer, panel
+//output: viewer result
+export async function spiralPlot(): Promise<DG.Viewer> {//(table: DG.DataFrame, activity: DG.Column) {
+// Read as dataframe
+  const table = await grok.data.files.openTable('Demo:TestJobs:Files:DemoFiles/bio/peptides.csv');
+  const activity = await table.columns.addNewCalculated('-log10(Activity)', '0-Log10(${Activity})');
+  view = grok.shell.addTableView(table);
+  return view.addViewer(SpiralPlot.fromTable(table, {valuesColumnName: activity.name}));
 }

package/src/peptides.ts

@@ -0,0 +1,93 @@
+import * as ui from 'datagrok-api/ui';
+import * as DG from 'datagrok-api/dg';
+import {createPeptideSimilaritySpaceViewer} from './utils/peptide-similarity-space';
+import {addViewerToHeader} from './viewers/stacked-barchart-viewer';
+
+/**
+ * Peptides controller class.
+ *
+ * @export
+ * @class Peptides
+ */
+export class Peptides {
+  /**
+   * Class initializer
+   *
+   * @param {DG.Grid} tableGrid Working talbe grid.
+   * @param {DG.TableView} view Working view.
+   * @param {DG.DataFrame} currentDf Working table.
+   * @param {{[key: string]: string}} options SAR viewer options
+   * @param {DG.Column} col Aligned sequences column.
+   * @param {string} activityColumnChoice Activity column name.
+   * @memberof Peptides
+   */
+  async init(
+    tableGrid: DG.Grid,
+    view: DG.TableView,
+    currentDf: DG.DataFrame,
+    options: {[key: string]: string},
+    col: DG.Column,
+    activityColumnChoice: string,
+  ) {
+    for (let i = 0; i < tableGrid.columns.length; i++) {
+      const aarCol = tableGrid.columns.byIndex(i);
+      if (aarCol &&
+          aarCol.name &&
+          aarCol.column?.semType != 'aminoAcids'
+      ) {
+        //@ts-ignore
+        tableGrid.columns.byIndex(i)?.visible = false;
+      }
+    }
+
+    const originalDfColumns = (currentDf.columns as DG.ColumnList).names();
+
+    const sarViewer = view.addViewer('peptide-sar-viewer', options);
+    const sarNode = view.dockManager.dock(sarViewer, DG.DOCK_TYPE.DOWN, null, 'SAR Viewer');
+
+    const sarViewerVertical = view.addViewer('peptide-sar-viewer-vertical');
+    view.dockManager.dock(sarViewerVertical, DG.DOCK_TYPE.RIGHT, sarNode, 'SAR Vertical Viewer');
+
+    const peptideSpaceViewer = await createPeptideSimilaritySpaceViewer(
+      currentDf,
+      col,
+      't-SNE',
+      'Levenshtein',
+      100,
+      view,
+      `${activityColumnChoice}Scaled`,
+    );
+    view.dockManager.dock(peptideSpaceViewer, DG.DOCK_TYPE.LEFT, sarNode, 'Peptide Space Viewer', 0.3);
+
+    const StackedBarchartProm = currentDf.plot.fromType('StackedBarChartAA');
+    addViewerToHeader(tableGrid, StackedBarchartProm);
+
+    const hideIcon = ui.iconFA('window-close', () => { //undo?, times?
+      const viewers = [];
+      for (const viewer of view.viewers) {
+        if (viewer.type !== DG.VIEWER.GRID) {
+          viewers.push(viewer);
+        }
+      }
+      viewers.forEach((v) => v.close());
+
+      const cols = (currentDf.columns as DG.ColumnList);
+      for (const colName of cols.names()) {
+        if (!originalDfColumns.includes(colName)) {
+          cols.remove(colName);
+        }
+      }
+
+      currentDf.selection.setAll(false);
+      currentDf.filter.setAll(true);
+
+      tableGrid.setOptions({'colHeaderHeight': 20});
+      tableGrid.columns.setVisible(originalDfColumns);
+
+      view.setRibbonPanels(ribbonPanels);
+    }, 'Close viewers and restore dataframe');
+
+    const ribbonPanels = view.getRibbonPanels();
+    view.setRibbonPanels([[hideIcon]]);
+  }
+}