@datagrok/bio 2.26.8 → 2.27.1

package/src/analysis/sequence-activity-cliffs.ts

@@ -181,12 +181,14 @@ export function createDifferencesWithPositions(
 }
 
 export function createLinesGrid(df: DG.DataFrame, colNames: string[]): DG.Grid {
-  const seqDiffCol = DG.Column.string('seq_diff', df.rowCount)
-    .init((i) => `${df.get(colNames[0], i)}#${df.get(colNames[1], i)}`);
-  seqDiffCol.semType = 'MacromoleculeDifference';
-  seqDiffCol.meta.units = df.col(colNames[0])!.meta.units;
-  seqDiffCol.setTag(bioTAGS.separator, df.col(colNames[0])!.getTag(bioTAGS.separator));
-  df.columns.add(seqDiffCol);
+  if (!df.col('seq_diff')) {
+    const seqDiffCol = DG.Column.string('seq_diff', df.rowCount)
+      .init((i) => `${df.get(colNames[0], i)}#${df.get(colNames[1], i)}`);
+    seqDiffCol.semType = 'MacromoleculeDifference';
+    seqDiffCol.meta.units = df.col(colNames[0])!.meta.units;
+    seqDiffCol.setTag(bioTAGS.separator, df.col(colNames[0])!.getTag(bioTAGS.separator));
+    df.columns.add(seqDiffCol);
+  }
   const grid = df.plot.grid();
   grid.col(colNames[0])!.visible = false;
   grid.col(colNames[1])!.visible = false;

package/src/package-api.ts

@@ -15,8 +15,8 @@ export namespace scripts {
   /**
   Converts molecules to HELM notation based on monomer library
   */
-  export async function molToHelmConverterPy(moleculesDataframe: DG.DataFrame , moleculesColumn: DG.Column , libraryJSON: string ): Promise<DG.DataFrame> {
-    return await grok.functions.call('Bio:MolToHelmConverterPy', { moleculesDataframe, moleculesColumn, libraryJSON });
+  export async function molToHelmConverterPy(moleculesDataframe: DG.DataFrame , moleculesColumn: DG.Column , libraryFile: DG.FileInfo ): Promise<DG.DataFrame> {
+    return await grok.functions.call('Bio:MolToHelmConverterPy', { moleculesDataframe, moleculesColumn, libraryFile });
   }
 
   /**
@@ -277,6 +277,13 @@ export namespace funcs {
     return await grok.functions.call('Bio:AlignSequences', { sequenceCol, clustersCol, options });
   }
 
+  /**
+  Aligns non-canonical peptide sequences using PepSeA Docker container (MAFFT)
+  */
+  export async function pepseaMsa(sequenceCol: DG.Column , method: string , gapOpen: number , gapExtend: number ): Promise<DG.Column> {
+    return await grok.functions.call('Bio:PepseaMsa', { sequenceCol, method, gapOpen, gapExtend });
+  }
+
   /**
   Visualizes sequence composition on a WebLogo plot
   */

package/src/package.g.ts

@@ -444,6 +444,18 @@ export async function alignSequences(sequenceCol: any, clustersCol: any, options
   return await PackageFunctions.alignSequences(sequenceCol, clustersCol, options);
 }
 
+//name: PepSeA
+//description: Aligns non-canonical peptide sequences using PepSeA Docker container (MAFFT)
+//input: column sequenceCol { semType: Macromolecule }
+//input: string method = 'mafft --auto' { choices: ["mafft --auto","mafft","linsi","ginsi","einsi","fftns","fftnsi","nwns","nwnsi"] }
+//input: double gapOpen = 1.53 
+//input: double gapExtend = 0 
+//output: column result
+//meta.role: sequenceMSA
+export async function pepseaMsa(sequenceCol: DG.Column<any>, method: string, gapOpen: number, gapExtend: number) : Promise<any> {
+  return await PackageFunctions.pepseaMsa(sequenceCol, method, gapOpen, gapExtend);
+}
+
 //name: Composition Analysis
 //description: Visualizes sequence composition on a WebLogo plot
 //output: viewer result

package/src/package.ts

@@ -33,6 +33,7 @@ import {getRdKitModule} from '@datagrok-libraries/bio/src/chem/rdkit-module';
 import {ISeqHandler, SeqTemps} from '@datagrok-libraries/bio/src/utils/macromolecule/seq-handler';
 import {MmcrTemps} from '@datagrok-libraries/bio/src/utils/cell-renderer-consts';
 
+import {checkCurrentView} from '@datagrok-libraries/utils/src/view-utils';
 import {getMacromoleculeColumns} from './utils/ui-utils';
 import {MacromoleculeDifferenceCellRenderer, MacromoleculeSequenceCellRenderer,} from './utils/cell-renderer';
 import {VdRegionsViewer} from './viewers/vd-regions-viewer';
@@ -58,6 +59,7 @@ import {demoToAtomicLevel} from './demo/bio03-atomic-level';
 import {checkInputColumnUI} from './utils/check-input-column';
 import {MsaWarning} from './utils/multiple-sequence-alignment';
 import {multipleSequenceAlignmentUI} from './utils/multiple-sequence-alignment-ui';
+import {alignWithPepsea, pepseaMethods} from './utils/pepsea';
 import {WebLogoApp} from './apps/web-logo-app';
 import {SplitToMonomersFunctionEditor} from './function-edtiors/split-to-monomers-editor';
 import {splitToMonomersUI} from './utils/split-to-monomers';
@@ -527,10 +529,7 @@ export class PackageFunctions {
     @grok.decorators.param({type: 'object', options: {optional: true}}) options?: (IUMAPOptions | ITSNEOptions) & Options,
     @grok.decorators.param({options: {optional: true}}) demo?: boolean): Promise<DG.Viewer | undefined> {
     //workaround for functions which add viewers to tableView (can be run only on active table view)
-    if (table.name !== grok.shell.tv.dataFrame.name) {
-      grok.shell.error(`Table ${table.name} is not a current table view`);
-      return;
-    }
+    checkCurrentView(table);
     if (!checkInputColumnUI(molecules, 'Activity Cliffs'))
       return;
 
@@ -556,6 +555,7 @@ export class PackageFunctions {
         axesNames: axesNames,
       }).call(undefined, undefined, {processed: false});
 
+      checkCurrentView(table);
       const view = grok.shell.tv;
 
       const description = `Molecules: ${molecules.name}, activities: ${activities.name}, method: ${methodName}, ${options ? `options: ${JSON.stringify(options)},` : ``} similarity: ${similarityMetric}, similarity cutoff: ${similarity}`;
@@ -733,10 +733,7 @@ export class PackageFunctions {
     @grok.decorators.param({options: {optional: true}}) isDemo?: boolean
   ): Promise<DG.ScatterPlotViewer | undefined> {
     //workaround for functions which add viewers to tableView (can be run only on active table view)
-    if (table.name !== grok.shell.tv.dataFrame.name) {
-      grok.shell.error(`Table ${table.name} is not a current table view`);
-      return;
-    }
+    checkCurrentView(table);
     if (!checkInputColumnUI(molecules, 'Sequence Space'))
       return;
     const clusterColName = table.columns.getUnusedName('Cluster (DBSCAN)');
@@ -755,6 +752,7 @@ export class PackageFunctions {
 
     let res: DG.ScatterPlotViewer | undefined;
     if (plotEmbeddings) {
+      checkCurrentView(table);
       const tv = grok.shell.tv;
       res = tv.scatterPlot({x: embedColsNames[0], y: embedColsNames[1], title: 'Sequence space'});
       const description = `Molecules column: ${molecules.name}, method: ${methodName}, ${options ? `options: ${JSON.stringify(options)},` : ``} similarity: ${similarityMetric}`;
@@ -812,7 +810,8 @@ export class PackageFunctions {
     // collect current monomer library
     const monomerLib = _package.monomerLib;
     const libJSON = JSON.stringify(monomerLib.toJSON());
-    await api.scripts.molToHelmConverterPy(table, molecules, libJSON);
+    const fileInfo = DG.FileInfo.fromString('monomerLib.json', libJSON);
+    await api.scripts.molToHelmConverterPy(table, molecules, fileInfo);
 
     // semtype is not automatically set, so we set it manually
     const newCol = table.columns.toList().find((c) => c.name.toLowerCase().includes('regenerated sequence') && c.semType !== DG.SEMTYPE.MACROMOLECULE);
@@ -978,6 +977,21 @@ export class PackageFunctions {
     return multipleSequenceAlignmentUI({col: sequenceCol, clustersCol: clustersCol, ...options}, _package.seqHelper);
   }
 
+  @grok.decorators.func({
+    name: 'PepSeA',
+    description: 'Aligns non-canonical peptide sequences using PepSeA Docker container (MAFFT)',
+    meta: {role: 'sequenceMSA'},
+    outputs: [{name: 'result', type: 'column'}],
+  })
+  static async pepseaMsa(
+    @grok.decorators.param({type: 'column', options: {semType: 'Macromolecule'}}) sequenceCol: DG.Column<string>,
+    @grok.decorators.param({type: 'string', options: {choices: ['mafft --auto', 'mafft', 'linsi', 'ginsi', 'einsi', 'fftns', 'fftnsi', 'nwns', 'nwnsi'], initialValue: 'mafft --auto'}}) method: string = 'mafft --auto',
+    @grok.decorators.param({type: 'double', options: {initialValue: '1.53'}}) gapOpen: number = 1.53,
+    @grok.decorators.param({type: 'double', options: {initialValue: '0'}}) gapExtend: number = 0,
+  ): Promise<DG.Column<string>> {
+    return alignWithPepsea(sequenceCol, method, gapOpen, gapExtend);
+  }
+
   @grok.decorators.func({
     name: 'Composition Analysis',
     description: 'Visualizes sequence composition on a WebLogo plot',

package/src/tests/msa-tests.ts

@@ -108,7 +108,7 @@ MWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHP
 
   async function _testMSAOnColumn(
     srcCsv: string, tgtCsv: string,
-    srcNotation: NOTATION, tgtNotation: NOTATION, alphabet?: ALPHABET, pepseaMethod?: string,
+    srcNotation: NOTATION, tgtNotation: NOTATION, alphabet?: ALPHABET, engineMethod?: string,
   ): Promise<void> {
     const srcDf: DG.DataFrame = DG.DataFrame.fromCsv(srcCsv);
     await grok.data.detectSemanticTypes(srcDf);
@@ -121,7 +121,11 @@ MWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHPMWRSWYCKHP
     if (alphabet)
       expect(srcCol.getTag(bioTAGS.alphabet), alphabet);
 
-    const msaSeqCol = await multipleSequenceAlignmentUI({col: srcCol, pepsea: {method: pepseaMethod}}, seqHelper);
+    const msaSeqCol = await multipleSequenceAlignmentUI({
+      col: srcCol,
+      engine: engineMethod ? 'PepSeA' : undefined,
+      engineParams: engineMethod ? {method: engineMethod} : undefined,
+    }, seqHelper);
     expect(msaSeqCol.semType, DG.SEMTYPE.MACROMOLECULE);
     expect(msaSeqCol.meta.units, tgtNotation);
     expect(msaSeqCol.getTag(bioTAGS.aligned), ALIGNMENT.SEQ_MSA);

package/src/utils/annotations/annotation-manager-ui.ts

@@ -63,7 +63,7 @@ export function showAnnotationManagerDialog(): void {
         setColumnAnnotations(selectedCol, updated);
         df.fireValuesChanged();
         refreshList();
-      });
+      }, 'Delete');
       removeBtn.style.cursor = 'pointer';
       removeBtn.style.color = '#999';
       removeBtn.style.marginLeft = '8px';

package/src/utils/constants.ts

@@ -1,5 +1,3 @@
-import {pepseaMethods} from './pepsea';
-
 export enum COLUMNS_NAMES {
   SPLIT_COL = '~split',
   ACTIVITY = '~activity',
@@ -64,14 +62,12 @@ export namespace PEPSEA {
 export const kalignVersion = '3.3.1';
 
 export const msaDefaultOptions = {
-  pepsea: {
-    gapOpen: 1.53,
-    gapExtend: 0,
-    method: pepseaMethods[0],
-  },
   kalign: {
     gapOpen: -1.0,
     gapExtend: -1.0,
     terminalGap: -1.0,
   },
 } as const;
+
+/** meta.role value for dynamically discovered MSA engine functions */
+export const MSA_ENGINE_ROLE = 'sequenceMSA';

package/src/utils/monomer-lib/library-file-manager/ui.ts

@@ -198,7 +198,7 @@ class LibraryControlsManager {
       updateLibrarySelectionStatus(libInput.value, libFileName);
     }});
     ui.tooltip.bind(libInput.root, `Include monomers from ${libFileName}`);
-    const deleteIcon = ui.iconFA('trash-alt', () => this.promptForLibraryDeletion(libFileName));
+    const deleteIcon = ui.iconFA('trash-alt', () => this.promptForLibraryDeletion(libFileName), 'Delete');
     const editIcon = ui.icons.edit(async () => {
       grok.shell.v = await (await MonomerManager.getInstance()).getViewRoot(libFileName);
     }, 'Edit monomer library');

package/src/utils/monomer-lib/monomer-manager/monomer-manager.ts

@@ -486,7 +486,7 @@ export class MonomerManager implements IMonomerManager {
       }
       const monomer = await monomerFromDfRow(this.tv!.dataFrame.rows.get(currentRowIdx));
       await this._newMonomerForm.removeMonomers([monomer], this.libInput.value!);
-    });
+    }, 'Delete');
 
     ui.tooltip.bind(deleteButton, () =>
       `${(this.tv?.dataFrame?.selection?.trueCount ?? 0) > 0 ? 'Delete selected monomers' : 'Delete monomer'}`);