@datagrok/bio 2.25.12 → 2.25.14

package/package.json

@@ -5,7 +5,7 @@
     "name": "Davit Rizhinashvili",
     "email": "drizhinashvili@datagrok.ai"
   },
-  "version": "2.25.12",
+  "version": "2.25.14",
   "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",

package/src/demo/bio03-atomic-level.ts

@@ -4,7 +4,6 @@ import * as DG from 'datagrok-api/dg';
 import {_package, PackageFunctions} from '../package';
 import {handleError} from './utils';
 import {DemoScript} from '@datagrok-libraries/tutorials/src/demo-script';
-import {delay} from '@datagrok-libraries/test/src/test';
 import {adjustGridcolAfterRender} from '../utils/ui-utils';
 
 export async function demoToAtomicLevel(): Promise<void> {
@@ -63,7 +62,7 @@ export async function demoBio03UI(): Promise<void> {
         dlg = ui.dialog()
           .add(sketcher)
           .show();
-        await delay(3000);
+        await DG.delay(3000);
         dlg.close();
       }, {
         description: 'Display atomic level structure within a sketcher.',

package/src/substructure-search/substructure-search.ts

@@ -2,7 +2,6 @@ import * as grok from 'datagrok-api/grok';
 import * as ui from 'datagrok-api/ui';
 import * as DG from 'datagrok-api/dg';
 
-import {delay} from '@datagrok-libraries/test/src/test';
 import {TAGS as bioTAGS, NOTATION} from '@datagrok-libraries/bio/src/utils/macromolecule';
 import {ISeqHelper} from '@datagrok-libraries/bio/src/utils/seq-helper';
 
@@ -154,7 +153,7 @@ export async function helmSubstructureSearch(
 export async function invalidateMols(col: DG.Column<string>, seqHelper: ISeqHelper, pattern: boolean): Promise<void> {
   const progressBar = DG.TaskBarProgressIndicator.create(`Invalidating molfiles for ${col.name}`);
   try {
-    await delay(10);
+    await DG.delay(10);
     const monomersDict = new Map();
     const monomericMolsCol = await getMonomericMols(col, seqHelper, pattern, monomersDict);
     col.temp[MONOMERIC_COL_TAGS.MONOMERIC_MOLS] = monomericMolsCol;

package/src/utils/detect-macromolecule-probe.ts

@@ -4,7 +4,6 @@ import * as DG from 'datagrok-api/dg';
 
 
 import {_package} from '../package';
-import {delay} from '@datagrok-libraries/test/src/test';
 
 type IDetectorReport = { categoriesSample: any[], rejectReason: string };
 
@@ -31,7 +30,7 @@ export async function detectMacromoleculeProbeDo(
       if ((progress - progressLast) >= 0.1) {
         progressLast = progress;
         pi.update(100 * progress, `detectMacromolecule probe ${failCount}/${i}/${probeCount} ...`);
-        await delay(0);
+        await DG.delay(0);
       }
     }
     if (failCount > 0)

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

@@ -4,7 +4,6 @@ import * as grok from 'datagrok-api/grok';
 import * as ui from 'datagrok-api/ui';
 import * as DG from 'datagrok-api/dg';
 
-import {delay} from '@datagrok-libraries/test/src/test';
 import {ILogger} from '@datagrok-libraries/bio/src/utils/logger';
 import {DEFAULT_FILES_LIB_PROVIDER_NAME, findProviderWithLibraryName, IMonomerLib, IMonomerSet} from '@datagrok-libraries/bio/src/types/monomer-library';
 import {
@@ -91,7 +90,7 @@ export class MonomerLibManager implements IMonomerLibHelper {
         return true;
       })(),
       (async () => {
-        await delay(timeout);
+        await DG.delay(timeout);
         return false;
       })(),
     ]).then((res) => {

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

@@ -0,0 +1,347 @@
+/* eslint-disable max-lines-per-function */
+/* eslint-disable max-len */
+import * as grok from 'datagrok-api/grok';
+import * as DG from 'datagrok-api/dg';
+
+import {IMonomerLib, Monomer} from '@datagrok-libraries/bio/src/types/monomer-library';
+import {PolymerType} from '@datagrok-libraries/bio/src/helm/types';
+
+import {STANDRARD_R_GROUPS} from './const';
+import {standardiseMonomers, capSmiles, getCorrectedSmiles} from './monomer-manager';
+
+/** Represents a single monomer match result — one monomer that matched a molecule */
+interface MonomerMatch {
+  /** monomer symbol from the library */
+  symbol: string;
+  /** canonical (possibly capped) SMILES used for the match */
+  smiles: string;
+  /** original SMILES from the monomer definition */
+  original: string;
+  /** library source name */
+  source: string;
+}
+
+/** Maps keyed by canonical SMILES, where each key can map to multiple monomers */
+type MonomerSmilesMap = {[smiles: string]: MonomerMatch[]};
+
+const MATCH_SEPARATOR = ' | ';
+
+/**
+ * Builds lookup maps from standardized monomers:
+ * - uncappedMap: maps raw canonical monomer SMILES -> MonomerMatch[]
+ * - cappedMap: maps capped (R-groups replaced with cap groups) canonical SMILES -> MonomerMatch[]
+ * Both maps store arrays so that duplicate monomers (same structure, different symbols/libs) are preserved.
+ */
+async function buildMonomerSmilesMaps(
+  fixedMonomers: Monomer[], originalMonomers: Monomer[], converterFunc: DG.Func,
+): Promise<{cappedMap: MonomerSmilesMap; uncappedMap: MonomerSmilesMap}> {
+  // build uncapped map from raw monomer SMILES
+  const uncappedMap: MonomerSmilesMap = {};
+  for (const m of fixedMonomers) {
+    if (!m.smiles) continue;
+    const match: MonomerMatch = {symbol: m.symbol, smiles: m.smiles, original: m.smiles, source: m.lib?.source ?? ''};
+    if (!uncappedMap[m.smiles]) uncappedMap[m.smiles] = [];
+    uncappedMap[m.smiles].push(match);
+  }
+
+  // build capped monomer entries: replace R-groups with cap group atoms
+  const cappedEntries = fixedMonomers
+    .map((m, i) => ({
+      symbol: m.symbol,
+      smiles: capSmiles(m.smiles ?? '', m.rgroups ?? []),
+      original: m.smiles,
+      source: originalMonomers[i]?.lib?.source ?? '',
+    }))
+    .filter((e) => !!e.smiles && !e.smiles.includes('[*:'));
+
+  // canonicalize all capped SMILES in bulk
+  const cappedSmilesCol = DG.Column.fromList(DG.COLUMN_TYPE.STRING, 'CappedSmiles', cappedEntries.map((e) => e.smiles));
+  cappedSmilesCol.semType = DG.SEMTYPE.MOLECULE;
+  const canonicalCappedCol: DG.Column = await converterFunc.apply({molecule: cappedSmilesCol, targetNotation: DG.chem.Notation.Smiles});
+  if (!canonicalCappedCol || canonicalCappedCol.length !== cappedSmilesCol.length)
+    throw new Error('Error canonicalizing capped monomer SMILES');
+
+  // build capped map with canonicalized SMILES as keys
+  const cappedMap: MonomerSmilesMap = {};
+  const canonicalCappedList = canonicalCappedCol.toList();
+  for (let i = 0; i < canonicalCappedList.length; i++) {
+    const smi = canonicalCappedList[i];
+    if (!smi) continue;
+    cappedEntries[i].smiles = smi;
+    const match: MonomerMatch = cappedEntries[i];
+    if (!cappedMap[smi]) cappedMap[smi] = [];
+    cappedMap[smi].push(match);
+  }
+
+  return {cappedMap, uncappedMap};
+}
+
+/**
+ * Corrects and canonicalizes the input molecule column.
+ * Handles both SMILES and molblock inputs.
+ * Returns the list of canonical SMILES strings (null for invalid molecules).
+ */
+async function canonicalizeMolecules(
+  molDf: DG.DataFrame, molColName: string, converterFunc: DG.Func,
+): Promise<(string | null)[]> {
+  const moleculesOriginalCol = molDf.col(molColName)!;
+  const correctedList = moleculesOriginalCol.toList().map((s) => {
+    if (!s) return s;
+    try {
+      const isMolBlock = s.includes('\n');
+      return getCorrectedSmiles([], isMolBlock ? undefined : s, isMolBlock ? s : undefined);
+    } catch (_e) {
+      return s;
+    }
+  });
+
+  const correctedCol = DG.Column.fromList(DG.COLUMN_TYPE.STRING, 'MoleculesCorrected', correctedList);
+  correctedCol.semType = DG.SEMTYPE.MOLECULE;
+  // dummy df needed for semtype detection by converterFunc
+  const _ddf = DG.DataFrame.fromColumns([correctedCol]);
+
+  const canonicalCol: DG.Column = await converterFunc.apply({molecule: correctedCol, targetNotation: DG.chem.Notation.Smiles});
+  if (!canonicalCol || canonicalCol.length !== correctedCol.length)
+    throw new Error('Error canonicalizing molecules');
+
+  return canonicalCol.toList();
+}
+
+/**
+ * Attempts to match a single canonical molecule SMILES against the lookup maps.
+ * Tries in order: capped map -> uncapped map -> cap the molecule with standard R-groups and retry.
+ * Returns all matching monomers (can be multiple from different libraries).
+ */
+function matchBySmiles(
+  canonicalMol: string, cappedMap: MonomerSmilesMap, uncappedMap: MonomerSmilesMap,
+): MonomerMatch[] {
+  // try direct lookup in capped and uncapped maps
+  let matches = cappedMap[canonicalMol] ?? uncappedMap[canonicalMol];
+  if (matches && matches.length > 0) return matches;
+
+  // fallback: cap the molecule with standard R-groups and try again
+  const cappedMol = capSmiles(canonicalMol, STANDRARD_R_GROUPS);
+  if (cappedMol !== canonicalMol) {
+    const correctedMol = grok.chem.convert(cappedMol, DG.chem.Notation.Unknown, DG.chem.Notation.Smiles);
+    matches = cappedMap[correctedMol] ?? uncappedMap[correctedMol];
+    if (matches && matches.length > 0) return matches;
+  }
+
+  return [];
+}
+
+/**
+ * Builds a Morgan fingerprint lookup map for all capped monomer SMILES.
+ * The map keys are fingerprint binary strings (via DG.BitSet.toBinaryString()),
+ * which allows fast exact matching that is tolerant of explicit hydrogen
+ * and minor stereochemistry differences.
+ */
+async function buildMonomerFingerprintMap(
+  cappedMap: MonomerSmilesMap,
+): Promise<{fpMap: {[fpString: string]: MonomerMatch[]}; cappedSmilesList: string[]}> {
+  const cappedSmilesList = Object.keys(cappedMap);
+  if (cappedSmilesList.length === 0)
+    return {fpMap: {}, cappedSmilesList: []};
+
+  const monomerCol = DG.Column.fromList(DG.COLUMN_TYPE.STRING, 'MonomerSmiles', cappedSmilesList);
+  monomerCol.semType = DG.SEMTYPE.MOLECULE;
+
+  const fpCol: DG.Column = await grok.functions.call('Chem:getMorganFingerprints', {molColumn: monomerCol});
+
+  const fpMap: {[fpString: string]: MonomerMatch[]} = {};
+  for (let i = 0; i < fpCol.length; i++) {
+    const fp: DG.BitSet | null = fpCol.get(i);
+    if (!fp) continue;
+    const fpStr = fp.toBinaryString();
+    // merge monomer matches from the SMILES map into the fingerprint map
+    const smilesMatches = cappedMap[cappedSmilesList[i]] ?? [];
+    if (!fpMap[fpStr]) fpMap[fpStr] = [];
+    fpMap[fpStr].push(...smilesMatches);
+  }
+
+  return {fpMap, cappedSmilesList};
+}
+
+/**
+ * For molecules that were not matched by exact SMILES, attempts matching via
+ * Morgan fingerprints. Computes fingerprints for unmatched molecules and looks
+ * them up in the monomer fingerprint map. Also tries capping with standard R-groups.
+ */
+async function matchByFingerprint(
+  unmatchedIndices: number[],
+  canonicalizedMolecules: (string | null)[],
+  monomerFpMap: {[fpString: string]: MonomerMatch[]},
+): Promise<Map<number, MonomerMatch[]>> {
+  const results = new Map<number, MonomerMatch[]>();
+  if (unmatchedIndices.length === 0 || Object.keys(monomerFpMap).length === 0)
+    return results;
+
+  // collect SMILES for unmatched molecules (uncapped first)
+  const uncappedSmiles: string[] = unmatchedIndices.map((idx) => canonicalizedMolecules[idx] ?? '');
+
+  // also prepare capped versions
+  const cappedSmiles: string[] = uncappedSmiles.map((s) => {
+    if (!s) return '';
+    const capped = capSmiles(s, STANDRARD_R_GROUPS);
+    return capped !== s ? grok.chem.convert(capped, DG.chem.Notation.Unknown, DG.chem.Notation.Smiles) : s;
+  });
+
+  // compute fingerprints for both uncapped and capped molecules
+  const uncappedCol = DG.Column.fromList(DG.COLUMN_TYPE.STRING, 'UnmatchedMols', uncappedSmiles);
+  uncappedCol.semType = DG.SEMTYPE.MOLECULE;
+  const cappedCol = DG.Column.fromList(DG.COLUMN_TYPE.STRING, 'UnmatchedMolsCapped', cappedSmiles);
+  cappedCol.semType = DG.SEMTYPE.MOLECULE;
+
+  const [uncappedFpCol, cappedFpCol]: [DG.Column, DG.Column] = await Promise.all([
+    grok.functions.call('Chem:getMorganFingerprints', {molColumn: uncappedCol}),
+    grok.functions.call('Chem:getMorganFingerprints', {molColumn: cappedCol}),
+  ]);
+
+  for (let i = 0; i < unmatchedIndices.length; i++) {
+    const molIdx = unmatchedIndices[i];
+
+    // try uncapped fingerprint first
+    const uncappedFp: DG.BitSet | null = uncappedFpCol.get(i);
+    if (uncappedFp) {
+      const fpStr = uncappedFp.toBinaryString();
+      const matches = monomerFpMap[fpStr];
+      if (matches && matches.length > 0) {
+        results.set(molIdx, matches);
+        continue;
+      }
+    }
+
+    // fallback: try capped fingerprint
+    const cappedFp: DG.BitSet | null = cappedFpCol.get(i);
+    if (cappedFp) {
+      const fpStr = cappedFp.toBinaryString();
+      const matches = monomerFpMap[fpStr];
+      if (matches && matches.length > 0)
+        results.set(molIdx, matches);
+    }
+  }
+
+  return results;
+}
+
+/** Deduplicates matches by symbol, keeping one entry per unique monomer symbol */
+function deduplicateMatches(matches: MonomerMatch[]): MonomerMatch[] {
+  const seen = new Set<string>();
+  return matches.filter((m) => {
+    if (seen.has(m.symbol)) return false;
+    seen.add(m.symbol);
+    return true;
+  });
+}
+
+/** Collects all source library names for matched monomers, including known duplicates */
+function collectSources(
+  matches: MonomerMatch[], duplicates: {[symbol: string]: Monomer[]},
+): string {
+  const sources = new Set<string>();
+  for (const m of matches) {
+    // check if monomerLib knows about duplicates for this symbol across libraries
+    const dups = duplicates[m.symbol];
+    if (dups && dups.length > 0) {
+      for (const dup of dups) {
+        const s = dup?.lib?.source;
+        if (s) sources.add(s);
+      }
+    } else if (m.source)
+      sources.add(m.source);
+  }
+  return Array.from(sources).join(', ');
+}
+
+/**
+ * Matches molecules in a dataframe with monomers from a monomer library.
+ *
+ * Matching pipeline:
+ *   1. Standardize monomers and build SMILES lookup maps (capped & uncapped)
+ *   2. Canonicalize input molecules
+ *   3. Phase 1: exact canonical SMILES matching (capped, uncapped, and fallback-capped molecule)
+ *   4. Phase 2: Morgan fingerprint fallback for molecules that didn't match by SMILES
+ *   5. Populate result columns (supports multiple matches per molecule via pipe-delimited values)
+ *
+ * @returns cloned input DataFrame with added match columns
+ */
+export async function matchMoleculesWithMonomers(
+  molDf: DG.DataFrame, molColName: string, monomerLib: IMonomerLib, polymerType: PolymerType = 'PEPTIDE',
+): Promise<DG.DataFrame> {
+  const duplicates = monomerLib.duplicateMonomers?.[polymerType] ?? {};
+  const converterFunc = DG.Func.find({package: 'Chem', name: 'convertMoleculeNotation'})[0];
+  if (!converterFunc)
+    throw new Error('Function convertMoleculeNotation not found, please install Chem package');
+
+  // === Step 1: Standardize monomers and build SMILES lookup maps ===
+  const monomers = monomerLib.getMonomerSymbolsByType(polymerType)
+    .map((s) => monomerLib.getMonomer(polymerType, s)!)
+    .filter((m) => m && (m.smiles || m.molfile));
+
+  const fixedMonomers = await standardiseMonomers(monomers);
+  // preserve library reference from original monomers (lost during standardization)
+  fixedMonomers.forEach((m, i) => { m.lib = monomers[i].lib; });
+
+  const {cappedMap, uncappedMap} = await buildMonomerSmilesMaps(fixedMonomers, monomers, converterFunc);
+
+  // === Step 2: Canonicalize input molecules ===
+  const canonicalizedMolecules = await canonicalizeMolecules(molDf, molColName, converterFunc);
+
+  // === Step 3: Phase 1 — Exact canonical SMILES matching ===
+  // matchResults[i] holds all MonomerMatch entries for molecule i (empty array if unmatched)
+  const matchResults: MonomerMatch[][] = new Array(canonicalizedMolecules.length).fill(null).map(() => []);
+  const unmatchedIndices: number[] = [];
+
+  for (let i = 0; i < canonicalizedMolecules.length; i++) {
+    const mol = canonicalizedMolecules[i];
+    if (!mol) continue;
+    const smilesMatches = matchBySmiles(mol, cappedMap, uncappedMap);
+    if (smilesMatches.length > 0)
+      matchResults[i] = smilesMatches;
+    else
+      unmatchedIndices.push(i);
+  }
+
+  // === Step 4: Phase 2 — Morgan fingerprint fallback for unmatched molecules ===
+  if (unmatchedIndices.length > 0) {
+    try {
+      const {fpMap} = await buildMonomerFingerprintMap(cappedMap);
+      const fpMatches = await matchByFingerprint(unmatchedIndices, canonicalizedMolecules, fpMap);
+      for (const [idx, matches] of fpMatches)
+        matchResults[idx] = matches;
+    } catch (e) {
+      console.warn('Fingerprint fallback matching failed, continuing with SMILES matches only:', e);
+    }
+  }
+
+  // === Step 5: Populate result columns ===
+  const resultDf = molDf.clone();
+  const symbolCol = resultDf.columns.addNewString(resultDf.columns.getUnusedName('Matched monomer symbol'));
+  const smilesCol = resultDf.columns.addNewString(resultDf.columns.getUnusedName('Matched monomer smiles'));
+  smilesCol.semType = DG.SEMTYPE.MOLECULE;
+  const sourceCol = resultDf.columns.addNewString(resultDf.columns.getUnusedName('Matched monomer source'));
+  const matchCountCol = resultDf.columns.addNewInt(resultDf.columns.getUnusedName('Match count'));
+  const matchMethodCol = resultDf.columns.addNewString(resultDf.columns.getUnusedName('Match method'));
+  resultDf.columns.setOrder([molColName, symbolCol.name, smilesCol.name, sourceCol.name, matchCountCol.name, matchMethodCol.name]);
+
+  for (let i = 0; i < matchResults.length; i++) {
+    const matches = matchResults[i];
+    if (matches.length === 0) continue;
+
+    // deduplicate matches by symbol (same monomer can appear from multiple lookup paths)
+    const uniqueMatches = deduplicateMatches(matches);
+
+    // collect all sources, including duplicates from the monomer library
+    const allSources = collectSources(uniqueMatches, duplicates);
+
+    symbolCol.set(i, uniqueMatches.map((m) => m.symbol).join(MATCH_SEPARATOR), false);
+    smilesCol.set(i, uniqueMatches[0].original ?? uniqueMatches[0].smiles, false);
+    sourceCol.set(i, allSources, false);
+    matchCountCol.set(i, uniqueMatches.length, false);
+    // fingerprint matches are those from phase 2 (indices that were in unmatchedIndices)
+    const method = unmatchedIndices.includes(i) ? 'fingerprint' : 'exact';
+    matchMethodCol.set(i, method, false);
+  }
+
+  return resultDf;
+}

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

@@ -15,11 +15,10 @@ import {PolymerType, MonomerType} from '@datagrok-libraries/bio/src/helm/types';
 import {MonomerLibManager} from '../lib-manager';
 
 import {MONOMER_RENDERER_TAGS} from '@datagrok-libraries/bio/src/utils/cell-renderer';
-import {BioTags} from '@datagrok-libraries/bio/src/utils/macromolecule/consts';
+import {BioTags, MONOMER_MOTIF_SPLITTER} from '@datagrok-libraries/bio/src/utils/macromolecule/consts';
 //@ts-ignore
 import '../../../../css/monomer-manager.css';
 import {Subscription} from 'rxjs';
-import {STANDRARD_R_GROUPS} from './const';
 
 // columns of monomers dataframe, note that rgroups is hidden and will be displayed as separate columns
 export enum MONOMER_DF_COLUMN_NAMES {
@@ -60,84 +59,7 @@ export async function standardiseMonomers(monomers: Monomer[]) {
   return fixedMonomers;
 }
 
-/// matches molecules in the dataframe with monomers in the library by canonical smiles
-export async function matchMoleculesWithMonomers(molDf: DG.DataFrame, molColName: string, monomerLib: IMonomerLib, polymerType: PolymerType = 'PEPTIDE'): Promise<DG.DataFrame> {
-  const duplicates = monomerLib.duplicateMonomers?.[polymerType] ?? {};
-  const converterFunc = DG.Func.find({package: 'Chem', name: 'convertMoleculeNotation'})[0];
-  if (!converterFunc)
-    throw new Error('Function convertMoleculeNotation not found, please install Chem package');
-  // first: stamdardize monomers
-  const monomers = monomerLib.getMonomerSymbolsByType(polymerType).map((s) => monomerLib.getMonomer(polymerType, s)!).filter((m) => m && (m.smiles || m.molfile));
-  const fixedMonomers = await standardiseMonomers(monomers);
-  fixedMonomers.forEach((m, i) => {
-    m.lib = monomers[i].lib;
-  });
-  const unCappedMonomerSmilesMap = fixedMonomers.filter((m) => !!m.smiles).reduce((acc, m) => {
-    acc[m.smiles] = {symbol: m.symbol, smiles: m.smiles, original: m.smiles, source: m.lib?.source}; return acc;
-  }, {} as {[smiles: string]: {symbol: string, smiles: string, original: string | undefined, source: string | undefined}});
-  const cappedMonomerSmiles = fixedMonomers.map((m, i) => ({symbol: m.symbol, smiles: capSmiles(m.smiles ?? '', m.rgroups ?? []), original: m.smiles, source: monomers[i]?.lib?.source}))
-    .filter((s) => !!s?.smiles && !s.smiles.includes('[*:'));
-
-  // canonicalize all monomer smiles
-  const monomerSmilesCol = DG.Column.fromList(DG.COLUMN_TYPE.STRING, 'MonomerSmiles', cappedMonomerSmiles.map((m) => m.smiles!));
-  monomerSmilesCol.semType = DG.SEMTYPE.MOLECULE;
-  const canonicalizedMonomersSmilesCol: DG.Column = await converterFunc.apply({molecule: monomerSmilesCol, targetNotation: DG.chem.Notation.Smiles});
-  if (!canonicalizedMonomersSmilesCol || canonicalizedMonomersSmilesCol.length !== monomerSmilesCol.length)
-    throw new Error('Error canonicalizing monomer smiles');
-  canonicalizedMonomersSmilesCol.toList().forEach((s, i) => cappedMonomerSmiles[i].smiles = s);
-  const cappedMonomerSmilesMap = cappedMonomerSmiles.reduce((acc, m) => { acc[m.smiles] = m; return acc; }, {} as {[smiles: string]: {symbol: string, smiles: string, original: string | undefined, source: string | undefined}});
-
-  const moleculesOriginalCol = molDf.col(molColName)!;
-  const correctedOriginalList = moleculesOriginalCol.toList().map((s) => {
-    if (!s) return s;
-    try {
-      const isMolBlock = s.includes('\n');
-      return getCorrectedSmiles([], isMolBlock ? undefined : s, isMolBlock ? s : undefined);
-    } catch (_e) {
-      return s;
-    }
-  });
-  const moleculesOriginalColCorrected = DG.Column.fromList(DG.COLUMN_TYPE.STRING, 'MoleculesOriginalCorrected', correctedOriginalList);
-  // create dummy df
-  moleculesOriginalColCorrected.semType = DG.SEMTYPE.MOLECULE;
-  const _ddf = DG.DataFrame.fromColumns([moleculesOriginalColCorrected]);
-  const canonicalizedMoleculesCol: DG.Column = await converterFunc.apply({molecule: moleculesOriginalColCorrected, targetNotation: DG.chem.Notation.Smiles});
-  if (!canonicalizedMoleculesCol || canonicalizedMoleculesCol.length !== moleculesOriginalColCorrected.length)
-    throw new Error('Error canonicalizing molecules');
-
-  const canonicalizedMolecules = canonicalizedMoleculesCol.toList();
-
-  const resultDf = molDf.clone();
-  const matchingMonomerSmilesCol = resultDf.columns.addNewString(resultDf.columns.getUnusedName('Matched monomer smiles'));
-  matchingMonomerSmilesCol.semType = DG.SEMTYPE.MOLECULE;
-  const matchingMonomerSymbolCol = resultDf.columns.addNewString(resultDf.columns.getUnusedName('Matched monomer symbol'));
-  matchingMonomerSymbolCol.semType = 'Monomer';
-  const sourceLibCol = resultDf.columns.addNewString(resultDf.columns.getUnusedName('Matched monomer source'));
-  resultDf.columns.setOrder([molColName, matchingMonomerSymbolCol.name, matchingMonomerSmilesCol.name, sourceLibCol.name]);
-
-  for (let i = 0; i < canonicalizedMolecules.length; i++) {
-    const mol = canonicalizedMolecules[i];
-    if (!mol) continue;
-    let match = cappedMonomerSmilesMap[mol] ?? unCappedMonomerSmilesMap[mol];
-    if (!match) {
-      // try capping the molecule and matching again
-      const cappedMol = capSmiles(mol, STANDRARD_R_GROUPS);
-      if (cappedMol !== mol) {
-        const correctedMol = grok.chem.convert(cappedMol, DG.chem.Notation.Unknown, DG.chem.Notation.Smiles);
-        match = cappedMonomerSmilesMap[correctedMol] ?? unCappedMonomerSmilesMap[correctedMol];
-      }
-    }
-    if (match) {
-      const matchSymbol = match.symbol;
-      const sources = (duplicates[matchSymbol]?.length ?? 0) > 0 ? duplicates[matchSymbol].map((m) => m?.lib?.source).filter((s) => !!s).join(', ') : (match.source ?? '');
-      const originalSmiles = match.original ?? match.smiles;
-      matchingMonomerSmilesCol.set(i, originalSmiles, false);
-      matchingMonomerSymbolCol.set(i, matchSymbol, false);
-      sourceLibCol.set(i, sources, false);
-    }
-  }
-  return resultDf;
-}
+export {matchMoleculesWithMonomers} from './match-molecules';
 
 /** Standardizes the monomer library
  * warning: throws error if the library is not valid or has invalid monomers
@@ -1217,7 +1139,7 @@ function replaceAllylsInSmiles(smiles: string): string {
 }
 
 /**NB! Can throw error */
-function getCorrectedSmiles(rgroups: RGroup[], smiles?: string, molBlock?: string): string {
+export function getCorrectedSmiles(rgroups: RGroup[], smiles?: string, molBlock?: string): string {
   if (smiles)
     smiles = replaceAllylsInSmiles(smiles);
   const isSmilesMalformed = !smiles || !grok.chem.checkSmiles(smiles);
@@ -1310,7 +1232,7 @@ export function getCorrectedMolBlock(molBlock: string) {
 }
 
 // reverse of r-group substitution, will substitute rgroups with cap groups
-function capSmiles(smiles: string, rgroups: RGroup[]) {
+export function capSmiles(smiles: string, rgroups: RGroup[]) {
   let newSmiles = smiles;
   rgroups.forEach((rg, i) => {
     const rgroupNum = rg.label[1] ?? `${i + 1}`; // if label is not in format R#, use index as number

package/src/utils/multiple-sequence-alignment-ui.ts

@@ -6,7 +6,6 @@ import * as DG from 'datagrok-api/dg';
 import * as ui from 'datagrok-api/ui';
 
 import {ColumnInputOptions} from '@datagrok-libraries/utils/src/type-declarations';
-import {delay} from '@datagrok-libraries/test/src/test';
 import {ALPHABET, NOTATION} from '@datagrok-libraries/bio/src/utils/macromolecule';
 import {ISeqHelper} from '@datagrok-libraries/bio/src/utils/seq-helper';
 
@@ -82,7 +81,7 @@ export async function multipleSequenceAlignmentUI(
         table: table, value: seqCol, onValueChanged: async (value: DG.Column<any>) => {
           if (!value || value.semType !== DG.SEMTYPE.MACROMOLECULE) {
             okBtn.disabled = true;
-            await delay(0); // to
+            await DG.delay(0); // to
             colInput.value = prevSeqCol as DG.Column<string>;
             return;
           }

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

@@ -50,9 +50,19 @@ export class SeqHandler implements ISeqHandler {
     if (col.type !== DG.TYPE.STRING)
       throw new Error(`Unexpected column type '${col.type}', must be '${DG.TYPE.STRING}'.`);
     this._column = col;
-    const units: string | null = this._column.meta.units;
-    if (!units)
-      throw new Error('Units are not specified in column');
+    let units: string | null = this._column.meta.units;
+    if (!units) {
+      // it may be from layout that the macromolecule semtype is set but every other tag is missing, so we manually run detectors
+      if (!this._column.temp['seqHandlerDetectorRun']) {
+        this._column.temp['seqHandlerDetectorRun'] = true;
+        const detectorFunc = DG.Func.find({name: 'detectMacromolecule', meta: {role: 'semTypeDetector'}})[0];
+        if (detectorFunc)
+          detectorFunc.applySync({col: this._column});
+        units = this._column.meta.units;
+      }
+      if (!units)
+        throw new Error('Units are not specified in column');
+    }
     this._units = units!;
 
     this._notation = this.getNotation();
@@ -182,7 +192,7 @@ export class SeqHandler implements ISeqHandler {
 
       let aligned = uh.column.getTag(TAGS.aligned);
       if (aligned == null) {
-        aligned = uh.stats.sameLength ? ALIGNMENT.SEQ_MSA : ALIGNMENT.SEQ;
+        aligned = uh.stats.sameLength || uh.column.categories.slice(0, 5).filter((a) => !!a).every((a) => a.length > 100) ? ALIGNMENT.SEQ_MSA : ALIGNMENT.SEQ;
         uh.column.setTag(TAGS.aligned, aligned);
       }
 

package/src/utils/split-to-monomers.ts

@@ -2,7 +2,6 @@ import * as grok from 'datagrok-api/grok';
 import * as ui from 'datagrok-api/ui';
 import * as DG from 'datagrok-api/dg';
 
-import {delay} from '@datagrok-libraries/test/src/test';
 import {checkInputColumnUI} from './check-input-column';
 import {splitAlignedSequences} from '@datagrok-libraries/bio/src/utils/splitter';
 import * as C from './constants';
@@ -17,7 +16,7 @@ export async function splitToMonomersUI(
 ): Promise<DG.DataFrame> {
   // Delay is required for initial function dialog to close before starting invalidating of molfiles.
   // Otherwise, dialog is freezing
-  await delay(10);
+  await DG.delay(10);
   if (!checkInputColumnUI(seqCol, 'Sequence space')) return table;
 
   const seqHelper = _package.seqHelper;

package/src/viewers/vd-regions-viewer.ts

@@ -4,7 +4,7 @@ import * as DG from 'datagrok-api/dg';
 
 import {fromEvent, Observable, Subject, Unsubscribable} from 'rxjs';
 
-import {testEvent} from '@datagrok-libraries/test/src/test';
+import {testEvent} from '@datagrok-libraries/utils/src/test';
 import {
   IVdRegionsViewer,
   VdRegion, VdRegionType,

package/src/viewers/web-logo-viewer.ts

@@ -22,7 +22,7 @@ import {
 import {errorToConsole} from '@datagrok-libraries/utils/src/to-console';
 import {intToHtmlA} from '@datagrok-libraries/utils/src/color';
 import {ISeqSplitted} from '@datagrok-libraries/bio/src/utils/macromolecule/types';
-import {testEvent} from '@datagrok-libraries/test/src/test';
+import {testEvent} from '@datagrok-libraries/utils/src/test';
 import {PromiseSyncer} from '@datagrok-libraries/bio/src/utils/syncer';
 import {GAP_SYMBOL} from '@datagrok-libraries/bio/src/utils/macromolecule/consts';
 import {IMonomerLibBase} from '@datagrok-libraries/bio/src/types/monomer-library';

package/src/widgets/bio-substructure-filter-helm.ts

@@ -9,7 +9,6 @@ import {App, IHelmWebEditor} from '@datagrok-libraries/bio/src/helm/types';
 import {getHelmHelper} from '@datagrok-libraries/bio/src/helm/helm-helper';
 import {ILogger} from '@datagrok-libraries/bio/src/utils/logger';
 import {errInfo} from '@datagrok-libraries/bio/src/utils/err-info';
-import {delay} from '@datagrok-libraries/test/src/test';
 import {ISeqHelper} from '@datagrok-libraries/bio/src/utils/seq-helper';
 
 import {updateDivInnerHTML} from '../utils/ui-utils';
@@ -141,7 +140,7 @@ export class HelmBioFilter extends BioFilterBase<BioFilterProps> /* implements I
     const logPrefix = `${this.viewerToLog()}.substructureSearch( column = <${column.name}> )`;
     _package.logger.debug(`${logPrefix}, start`);
     try {
-      await delay(10);
+      await DG.delay(10);
       const res = await helmSubstructureSearch(this.props.substructure, column, this.seqHelper);
       return res;
     } finally {
@@ -163,7 +162,7 @@ export class HelmBioFilter extends BioFilterBase<BioFilterProps> /* implements I
   // async awaitRendered(timeout: number = 10000): Promise<void> {
   //   const callLog = `awaitRendered( ${timeout} )`;
   //   const logPrefix = `${this.viewerToLog()}.${callLog}`;
-  //   await delay(0);
+  //   await DG.delay(0);
   //   await testEvent(this.onRendered, () => {
   //     this.logger.debug(`${logPrefix}, ` + '_onRendered event caught');
   //   }, () => {