@datagrok/bio 2.21.9 → 2.21.11

package/package.json

@@ -5,7 +5,7 @@
     "name": "Leonid Stolbov",
     "email": "lstolbov@datagrok.ai"
   },
-  "version": "2.21.9",
+  "version": "2.21.11",
   "description": "Bioinformatics support (import/export of sequences, conversion, visualization, analysis). [See more](https://github.com/datagrok-ai/public/blob/master/packages/Bio/README.md) for details.",
   "repository": {
     "type": "git",
@@ -44,7 +44,7 @@
   ],
   "dependencies": {
     "@biowasm/aioli": "^3.1.0",
-    "@datagrok-libraries/bio": "^5.53.0",
+    "@datagrok-libraries/bio": "^5.53.2",
     "@datagrok-libraries/chem-meta": "^1.2.7",
     "@datagrok-libraries/math": "^1.2.4",
     "@datagrok-libraries/ml": "^6.10.0",

package/src/package.ts

@@ -462,6 +462,11 @@ export async function activityCliffs(table: DG.DataFrame, molecules: DG.Column<s
   similarity: number, methodName: DimReductionMethods,
   similarityMetric: MmDistanceFunctionsNames | BitArrayMetrics, preprocessingFunction: DG.Func,
   options?: (IUMAPOptions | ITSNEOptions) & Options, 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 an current table view`);
+    return;
+  }
   if (!checkInputColumnUI(molecules, 'Activity Cliffs'))
     return;
   const axesNames = getEmbeddingColsNames(table);
@@ -588,6 +593,11 @@ export async function sequenceSpaceTopMenu(table: DG.DataFrame, molecules: DG.Co
   plotEmbeddings: boolean, preprocessingFunction?: DG.Func, options?: (IUMAPOptions | ITSNEOptions) & Options,
   clusterEmbeddings?: boolean, 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 an current table view`);
+    return;
+  }
   const tableView =
     grok.shell.tv.dataFrame == table ? grok.shell.tv : undefined;
   if (!checkInputColumnUI(molecules, 'Sequence Space'))

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

@@ -194,6 +194,10 @@ export class MonomerManager implements IMonomerManager {
         args.menu.item('Edit Monomer', async () => {
           await this.editMonomer(this.tv!.dataFrame.rows.get(rowIdx));
         });
+
+        args.menu.item('Fix all monomers', () => {
+          this.fixAllMonomers();
+        });
         if (this.tv!.dataFrame.selection.trueCount > 0) {
           args.menu.item('Remove Selected Monomers', async () => {
             const monomers = await Promise.all(Array.from(this.tv!.dataFrame.selection.getSelectedIndexes())
@@ -271,6 +275,8 @@ export class MonomerManager implements IMonomerManager {
       this._newMonomerForm.setEmptyMonomer();
     }, 'Add New Monomer');
 
+    const fixAllMonomersIcon = ui.iconFA('wand-magic', () => { this.fixAllMonomers(); }, 'Fix all monomers');
+
     const editButton = ui.icons.edit(async () => {
       if ((this.tv?.dataFrame?.currentRowIdx ?? -1) < 0) return;
       await this.editMonomer(this.tv!.dataFrame.rows.get(this.tv!.dataFrame.currentRowIdx));
@@ -309,7 +315,7 @@ export class MonomerManager implements IMonomerManager {
       DG.Utils.download(libName!, lib!, 'text/plain');
     }, 'Download Monomer Library');
 
-    ribbons.push([newMonomerButton, editButton, deleteButton, downloadButton]);
+    ribbons.push([newMonomerButton, editButton, fixAllMonomersIcon, deleteButton, downloadButton]);
     this.tv.setRibbonPanels(ribbons);
 
 
@@ -443,6 +449,36 @@ export class MonomerManager implements IMonomerManager {
     }
   }
 
+  async fixAllMonomers() {
+    ui.dialog('Fix All Monomers')
+      .add(ui.divText('This action will fix all monomers in the library, standardize their smiles, molblocks and r-groups, assign correct natural analogs and save the library.'))
+      .add(ui.divText('Do you wish to continue?'))
+      .onOK(async () => {
+        const monomerDf = this.tv?.dataFrame;
+        let libName = this.libInput.value;
+        if (!monomerDf || !libName) {
+          grok.shell.error('No monomer library loaded');
+          return;
+        }
+        this.tv?.grid && ui.setUpdateIndicator(this.tv.grid.root, true);
+        try {
+          const monomers = await Promise.all(new Array(monomerDf.rowCount).fill(0).map((_, i) => monomerFromDfRow(monomerDf.rows.get(i))));
+          const monomersString = JSON.stringify(monomers.map((m) => ({...m, lib: undefined, wem: undefined})), null, 2);
+          if (!libName.endsWith('.json'))
+            libName += '.json';
+          await grok.dapi.files.writeAsText(LIB_PATH + libName, monomersString);
+          await this.monomerLibManamger.loadLibraries(true);
+          //await this.monomerLibManamger.loadLibraries(true);
+          grok.shell.v = await this.getViewRoot(libName);
+        } catch (e) {
+          grok.shell.error('Error saving library');
+          console.error(e);
+        } finally {
+          this.tv?.grid && ui.setUpdateIndicator(this.tv.grid.root, false);
+        }
+      }).show();
+  }
+
   public resetCurrentRowFollowing() {
     this._newMonomerForm.molChanged = false;
   }
@@ -1056,9 +1092,14 @@ function getCorrectedMolBlock(molBlock: string) {
     lines[isoLineIdx] = lines[isoLineIdx].substring(0, isoIndex) + 'RGP' + lines[isoLineIdx].substring(isoIndex + 3);
   }
 
-  const molStartIdx = lines.findIndex((line) => line.includes('V2000') || line.includes('V3000'));
+  const molStartIdx = lines.findIndex((line) => line.includes('V2000'));
 
-  const atomCount = Number.parseInt(lines[molStartIdx].trim().split(' ')[0]);
+  if (molStartIdx === -1) {
+    console.error('Mol start line not found');
+    return molBlock;
+  }
+  // only 3 positions are used for atom count, so we can safely parse it
+  const atomCount = Number.parseInt(lines[molStartIdx].trim().split(' ')[0].slice(0, 3).trim());
   const rgroupLineNumbers: { [atomLine: number]: number } = {};
   for (let atomI = molStartIdx + 1; atomI < molStartIdx + 1 + atomCount; atomI++) {
     const rIdx = lines[atomI].indexOf('R ');

package/src/utils/seq-helper/seq-handler.ts

@@ -20,8 +20,6 @@ import {ConvertFunc, ISeqHandler, JoinerFunc, SeqTemps, SeqValueBase} from '@dat
 
 import {SeqHelper} from './seq-helper';
 
-/* eslint-enable max-len */
-
 /** Class for handling notation units in Macromolecule columns and
  * conversion of notation systems in Macromolecule columns
  */
@@ -322,6 +320,44 @@ export class SeqHandler implements ISeqHandler {
     }
   }
 
+  /// Faster method to get monomers at certain position.
+  /// for canonical sequences in fasta (large proteins/nucleotides)
+  /// will be faster than getSplitted(rowIdx).getCanonical(posIdx)
+  getMonomerAtPosition(rowIdx: number, posIdx: number, canonical: boolean): string {
+    if (this.isCanonicalAlphabet && this.isFasta() && !this.getAlphabetIsMultichar()) {
+      const seq = this.column.get(rowIdx) ?? '';
+      const res = seq[posIdx];
+      return canonical ? (res === GapOriginals[NOTATION.FASTA] ? GAP_SYMBOL : (res ?? GAP_SYMBOL)) : (res ?? '');
+    }
+    const mSeq: ISeqSplitted = this.getSplitted(rowIdx);
+    if (posIdx < 0 || posIdx >= mSeq.length)
+      return this.defaultGapOriginal;
+    return canonical ? mSeq.getCanonical(posIdx) : mSeq.getOriginal(posIdx);
+  }
+
+  getMonomersAtPosition(position: number, canonical: boolean): string[] {
+    const length = this.column.length;
+    const res: string[] = new Array(length).fill(GAP_SYMBOL);
+    if (this.isCanonicalAlphabet && this.isFasta() && !this.getAlphabetIsMultichar()) {
+      const colCategories = this.column.categories;
+      const colIndexes = this.column.getRawData();
+      for (let i = 0; i < length; i++) {
+        const seq = colCategories[colIndexes[i]] ?? '';
+        if (position < seq.length) {
+          const resChar = seq[position];
+          res[i] = canonical ? (resChar === GapOriginals[NOTATION.FASTA] ? GAP_SYMBOL : (resChar ?? GAP_SYMBOL)) : (resChar ?? '');
+        }
+      }
+    } else {
+      for (let i = 0; i < length; i++) {
+        const mSeq: ISeqSplitted = this.getSplitted(i);
+        if (position >= 0 && position < mSeq.length)
+          res[i] = canonical ? mSeq.getCanonical(position) : mSeq.getOriginal(position);
+      }
+    }
+    return res;
+  }
+
   /** Any Macromolecule can be represented on Helm format. The reverse is not always possible. */
   public getValue(rowIdx: number, options?: any): SeqValueBase {
     const seq: string = this.column.get(rowIdx);
@@ -344,6 +380,11 @@ export class SeqHandler implements ISeqHandler {
 
   private _stats: SeqColStats | null = null;
 
+  public get isCanonicalAlphabet(): boolean {
+    const alphabet = this.alphabet;
+    return alphabet === ALPHABET.DNA || alphabet === ALPHABET.RNA || alphabet === ALPHABET.PT;
+  }
+
   public get stats(): SeqColStats {
     if (this._stats === null) {
       const freq: { [m: string]: number } = {};
@@ -393,6 +434,8 @@ export class SeqHandler implements ISeqHandler {
 
   public isSeparator(): boolean { return this.notation === NOTATION.SEPARATOR || !!this.separator; }
 
+  public isFastaOrSeparator(): boolean { return this.isFasta() || this.isSeparator(); }
+
   public isHelm(): boolean { return this.notation === NOTATION.HELM; }
 
   public isCustom(): boolean { return this.notation === NOTATION.CUSTOM; }

package/src/widgets/sequence-scrolling-widget.ts

@@ -51,10 +51,6 @@ export function handleSequenceHeaderRendering() {
         const split = sh.splitter(seq);
         const maxSeqLen = split ? split.length : 30;
 
-        // makes no sense to have scroller if we have shorter than 50 positions
-        if (maxSeqLen < 50)
-          continue;
-
         const defaultHeaderHeight = 40;
         const scroller = new MSAScrollingHeader({
           canvas: grid.overlay,
@@ -77,7 +73,10 @@ export function handleSequenceHeaderRendering() {
             });
           },
         });
-        grid.props.colHeaderHeight = 65;
+        // adjust header hight automatically only if the sequences are long enough
+        if (maxSeqLen > 40)
+          grid.props.colHeaderHeight = 65;
+
         setTimeout(() => { if (grid.isDetached) return; gCol.width = 400; }, 300); // needed because renderer sets its width
         grid.sub(grid.onCellRender.subscribe((e) => {
           const cell = e.cell;