@datagrok/bio 2.22.10 → 2.22.12

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

@@ -57,6 +57,53 @@ 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 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);
+  const cappedSmilse = fixedMonomers.map((m, i) => ({sym: 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', cappedSmilse.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) => cappedSmilse[i].smiles = s);
+
+  const molecules = molDf.col(molColName)!;
+  const canonicalizedMoleculesCol: DG.Column = await converterFunc.apply({molecule: molecules, targetNotation: DG.chem.Notation.Smiles});
+  if (!canonicalizedMoleculesCol || canonicalizedMoleculesCol.length !== molecules.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;
+    for (let j = 0; j < cappedSmilse.length; j++) {
+      if (cappedSmilse[j].smiles === mol) {
+        matchingMonomerSmilesCol.set(i, cappedSmilse[j].original!, false);
+        matchingMonomerSymbolCol.set(i, cappedSmilse[j].sym, false);
+        sourceLibCol.set(i, cappedSmilse[j].source ?? '', false);
+        break;
+      }
+    }
+  }
+  return resultDf;
+}
+
 /** Standardizes the monomer library
  * warning: throws error if the library is not valid or has invalid monomers
  */
@@ -127,7 +174,7 @@ export function getMonomersDataFrame(monomers: Monomer[]) {
           monomers[i].id,
           JSON.stringify(monomers[i].meta ?? {}),
           monomers[i].lib?.source ?? '',
-        ]);
+        ], false);
         // something is wrong with setting dates, so setting it manually for now
         try {
           if (date)

package/src/utils/save-as-fasta.ts

@@ -15,23 +15,21 @@ const FASTA_LINE_WIDTH = 60;
 export function saveAsFastaUI(): void {
   // Use grid for column order adjusted by user
   const grid: DG.Grid = grok.shell.tv.grid;
+  const dataFrame: DG.DataFrame = grid.dataFrame;
 
-  const idGColList: DG.GridColumn[] = wu.count(0).take(grid.columns.length)
-    .map((colI: number) => grid.columns.byIndex(colI)!)
-    .filter((gcol: DG.GridColumn) => gcol.column ? gcol.column.semType !== DG.SEMTYPE.MACROMOLECULE : false).toArray();
-  const defaultIdGCol: DG.GridColumn | undefined = idGColList
-    .find((gcol: DG.GridColumn) => gcol.name.toLowerCase().indexOf('id') !== -1);
-  const idDefaultValue = defaultIdGCol ? [defaultIdGCol.name] : [];
+  const idGColList: DG.Column[] = wu.count(0).take(dataFrame.columns.length)
+    .map((colI: number) => dataFrame.columns.byIndex(colI)!)
+    .filter((col: DG.Column) => col.semType !== DG.SEMTYPE.MACROMOLECULE).toArray();
+  const defaultIdGCol: DG.Column | undefined = idGColList
+    .find((col: DG.Column) => col.name.toLowerCase().indexOf('id') !== -1);
+  const idDefaultValue = defaultIdGCol ? [defaultIdGCol] : [];
 
-  const idGColListInput = ui.input.multiChoice('Seq id columns', {
-    value: idDefaultValue,
-    items: idGColList.map((gcol: DG.GridColumn) => gcol.name)
-  });
+  const idGColListInput = ui.input.columns('Seq id columns', {table: dataFrame, value: idDefaultValue,
+    filter: (col: DG.Column) => col.semType !== DG.SEMTYPE.MACROMOLECULE});
 
-  const seqGColList: DG.GridColumn[] = wu.count(0).take(grid.columns.length)/* range rom 0 to grid.columns.length */
-    .map((colI: number) => grid.columns.byIndex(colI)!)
-    .filter((gc: DG.GridColumn) => {
-      const col: DG.Column | null = gc.column;
+  const seqGColList: DG.Column[] = wu.count(0).take(dataFrame.columns.length)/* range rom 0 to grid.columns.length */
+    .map((colI: number) => dataFrame.columns.byIndex(colI)!)
+    .filter((col: DG.Column) => {
       if (col && col.semType === DG.SEMTYPE.MACROMOLECULE) {
         const sh = _package.seqHelper.getSeqHandler(col);
         return sh.isFasta();
@@ -39,10 +37,17 @@ export function saveAsFastaUI(): void {
       return false;
     }).toArray();
 
-  const seqDefaultValue = seqGColList.length > 0 ? seqGColList[0].name : [];
-  const seqColInput = ui.input.choice('Seq column', {
-    value: seqDefaultValue,
-    items: seqGColList.map((gCol: DG.GridColumn) => gCol.name)
+  const seqDefaultValue = seqGColList.length > 0 ? seqGColList[0] : null;
+  const seqColInput = ui.input.column('Seq column', {
+    table: dataFrame,
+    value: seqDefaultValue!,
+    filter: (col) => {
+      if (col && col.semType === DG.SEMTYPE.MACROMOLECULE) {
+        const sh = _package.seqHelper.getSeqHandler(col);
+        return sh.isFasta();
+      }
+      return false;
+    }
   });
 
   const lineWidthInput = ui.input.int('FASTA line width', {value: FASTA_LINE_WIDTH});
@@ -54,10 +59,8 @@ export function saveAsFastaUI(): void {
       lineWidthInput,
     ]))
     .onOK(() => {
-      const valueIdColList: DG.Column[] = idGColListInput.value ?
-        idGColListInput.value.map((colName: string) => grid.columns.byName(colName)!.column!) : [];
-      const valueSeqCol: DG.Column | null = seqColInput.value ?
-        grid.columns.byName(seqColInput.value as string)!.column : null;
+      const valueIdColList: DG.Column[] = idGColListInput.value ?? [];
+      const valueSeqCol: DG.Column | null = seqColInput.value ?? null;
       const valueLineWidth = lineWidthInput.value ?? FASTA_LINE_WIDTH;
 
       if (!valueSeqCol)