@datagrok/bio 2.20.4 → 2.21.0

package/files/tests/libraries/HELMmonomerSchema.json

@@ -85,12 +85,10 @@
     "symbol",
     "name",
     "molfile",
-    "author",
     "id",
     "rgroups",
     "smiles",
     "polymerType",
-    "monomerType",
-    "createDate"
+    "monomerType"
   ]
 }

package/package.json

@@ -5,7 +5,7 @@
     "name": "Leonid Stolbov",
     "email": "lstolbov@datagrok.ai"
   },
-  "version": "2.20.4",
+  "version": "2.21.0",
   "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",
@@ -65,7 +65,7 @@
     "wu": "^2.1.0"
   },
   "devDependencies": {
-    "@datagrok-libraries/helm-web-editor": "^1.1.13",
+    "@datagrok-libraries/helm-web-editor": "^1.1.14",
     "@datagrok-libraries/js-draw-lite": "^0.0.10",
     "@datagrok/chem": "^1.13.0",
     "@datagrok/dendrogram": "^1.2.33",

package/src/package.ts

@@ -543,7 +543,6 @@ export async function macromoleculePreprocessingFunction(
 }
 
 //name: Helm Fingerprints
-//tags: dim-red-preprocessing-function
 //meta.supportedSemTypes: Macromolecule
 //meta.supportedTypes: string
 //meta.supportedUnits: helm
@@ -581,7 +580,6 @@ export async function helmPreprocessingFunction(
 //input: object options {optional: true}
 //input: bool clusterEmbeddings = true { optional: true }
 //input: bool isDemo {optional: true}
-//output: viewer result
 //editor: Bio:SequenceSpaceEditor
 export async function sequenceSpaceTopMenu(table: DG.DataFrame, molecules: DG.Column,
   methodName: DimReductionMethods, similarityMetric: BitArrayMetrics | MmDistanceFunctionsNames,

package/src/tests/sequence-space-utils.ts

@@ -11,6 +11,7 @@ export async function _testSequenceSpaceReturnsResult(
 ) {
   // await grok.data.detectSemanticTypes(df);
   const col: DG.Column = df.getCol(colName);
+  df.name = 'seqSpaceDf';
   const semType: string = await grok.functions.call('Bio:detectMacromolecule', {col: col});
   if (semType)
     col.semType = semType;
@@ -18,12 +19,14 @@ export async function _testSequenceSpaceReturnsResult(
   const preprocessingFunc = DG.Func.find({package: 'Bio', name: 'macromoleculePreprocessingFunction'})[0];
   if (!preprocessingFunc)
     throw new Error('Preprocessing function not found');
-  const sp = await grok.functions.call('Bio:sequenceSpaceTopMenu', {
+  await grok.functions.call('Bio:sequenceSpaceTopMenu', {
     table: df, molecules: df.col(colName)!,
     methodName: algorithm, similarityMetric: MmDistanceFunctionsNames.LEVENSHTEIN,
     plotEmbeddings: true, preprocessingFunction: preprocessingFunc, options: {[BYPASS_LARGE_DATA_WARNING]: true}
   });
   // const sp = await sequenceSpaceTopMenu(df, df.col(colName)!, algorithm, MmDistanceFunctionsNames.LEVENSHTEIN, true,
   //   preprocessingFunc, {[BYPASS_LARGE_DATA_WARNING]: true});
-  expect(sp != null, true);
+  const tv = grok.shell.tableView(df.name);
+  const sp = Array.from(tv?.viewers ?? [])[1];
+  expect(sp != null);
 }

package/src/utils/cell-renderer.ts

@@ -31,6 +31,7 @@ import {CellRendererWithMonomerLibBackBase} from './monomer-cell-renderer-base';
 import * as C from './constants';
 
 import {_package} from '../package';
+import {SeqTemps} from '@datagrok-libraries/bio/src/utils/macromolecule/seq-handler';
 
 type TempType = { [tagName: string]: any };
 
@@ -221,7 +222,7 @@ export class MacromoleculeDifferenceCellRendererBack extends CellRendererWithMon
     w = getUpdatedWidth(grid, g, x, w, dpr);
     //TODO: can this be replaced/merged with splitSequence?
     const [s1, s2] = s.split('#');
-    const splitter = getSplitter(units, separator);
+    const splitter = this.tableCol.temp[SeqTemps.notationProvider]?.separatorSplitter ?? getSplitter(units, separator);
     const s1SS = splitter(s1);
     const s2SS = splitter(s2);
     const subParts1 = wu.count(0).take(s1SS.length).map((posIdx) => s1SS.getCanonical(posIdx)).toArray();

package/src/utils/helm-to-molfile/converter/converter.ts

@@ -109,7 +109,7 @@ export class HelmToMolfileConverter implements IHelmToMolfileConverter {
   public convertToMolfileV3KColumn(helmCol: DG.Column<string>): DG.Column<string> {
     const df = helmCol.dataFrame;
     const molfileList = this.convertToMolfileV3K(helmCol.toList()).map((mwm) => mwm.molfile);
-    const molColName = getUnusedColName(df, `molfileV2K(${helmCol.name})`);
+    const molColName = getUnusedColName(df, `molfileV3K(${helmCol.name})`);
     const molfileColumn = DG.Column.fromList('string', molColName, molfileList);
     return molfileColumn;
   }

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

@@ -54,7 +54,7 @@ export class MonomerLibFileValidator {
           `Bio: Monomer Library File Validator file ${fileName}, monomer '${name}' violating JSON schema:`,
           monomer,
           '\nError reason: ',
-          this.validateMonomerSchema.errors,
+          JSON.stringify(this.validateMonomerSchema.errors ?? {}),
           `\nThere may be other errors in ${fileName} since the validation is stopped after the first error.`,
           ' Please, verify that the monomer library file satisfies the JSON schema'
         );

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

@@ -54,7 +54,7 @@ export class SeqHandler implements ISeqHandler {
     const defaultGapOriginal = this.isFasta() ? GapOriginals[NOTATION.FASTA] :
       this.isSeparator() ? GapOriginals[NOTATION.SEPARATOR] :
         this.isHelm() ? GapOriginals[NOTATION.HELM] :
-          this.isCustom() ? this.notationProvider.defaultGapOriginal :
+          this.isCustom() ? (this.notationProvider?.defaultGapOriginal ?? GapOriginals[NOTATION.SEPARATOR]) :
             undefined;
     if (defaultGapOriginal == undefined)
       throw new Error(`Unexpected defaultGapOriginal for notation '${this.notation}'`);
@@ -97,6 +97,72 @@ export class SeqHandler implements ISeqHandler {
     }
 
     this.columnVersion = this.column.version;
+    // refine separator only at this stage
+    if (this.isSeparator() && (!this.isCustom() || !this.notationProvider) && !col.temp['seqHandlerRefined']) {
+      this.refineSeparatorNotation();
+      col.temp['seqHandlerRefined'] = true;
+    }
+  }
+
+  private async refineSeparatorNotation() {
+    // from detectors.
+    const SEQ_SAMPLE_LIMIT = 100; const SEQ_SAMPLE_LENGTH_LIMIT = 100;
+    const categoriesSample = [...new Set((this.column.length < SEQ_SAMPLE_LIMIT ?
+      wu.count(0).take(Math.min(SEQ_SAMPLE_LIMIT, this.column.length)).map((rowI) => this.column.get(rowI)) :
+      this.column.categories.slice(0, SEQ_SAMPLE_LIMIT))
+      .map((seq) => !!seq ? seq.substring(0, SEQ_SAMPLE_LENGTH_LIMIT * 5) : '')
+      .filter((seq) => seq.length !== 0/* skip empty values for detector */),
+    )];
+
+    const getStats = (values: string[], minLength: number, splitter: (s: string) => string[]) => {
+      const freq:{[key: string]: number} = {};
+      let sameLength = true;
+      let firstLength = null;
+
+      for (const seq of values) {
+        const mSeq = !!seq ? splitter(seq) : [];
+
+        if (firstLength === null)
+          firstLength = mSeq.length;
+        else if (mSeq.length !== firstLength)
+          sameLength = false;
+
+        if (mSeq.length >= minLength) {
+          for (const m of mSeq) {
+            if (!(m in freq)) freq[m] = 0;
+            freq[m] += 1;
+          }
+        }
+      }
+      return {freq: freq, sameLength: sameLength};
+    };
+
+    const stats = getStats(categoriesSample, 3, (s) => s.split(this.separator!));
+    let invalidateRequired = false;
+
+    const refinerList = [
+      {package: 'SequenceTranslator', name: 'refineNotationProviderForHarmonizedSequence'},
+    ];
+
+    for (const refineFuncFind of refinerList) {
+      try {
+        const funcList = DG.Func.find(refineFuncFind);
+        if (funcList.length === 0) continue;
+
+        const funcFc = funcList[0].prepare({col: this.column, stats: stats, separator: this.separator});
+        const refineRes = (await funcFc.call()).getOutputParamValue();
+        invalidateRequired ||= refineRes;
+      } catch (err) {
+        console.error(err);
+      }
+    }
+
+    if (invalidateRequired) {
+    // Applying custom notation provider MUST invalidate SeqHandler
+      delete this.column.temp[SeqTemps.seqHandler];
+
+      this.column.fireValuesChanged();
+    }
   }
 
   /** From detectMacromolecule */

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

@@ -89,11 +89,16 @@ export class SeqHelper implements ISeqHelper {
     //#endregion From HelmToMolfileConverter
     const helmList = helmCol.toList();
     const molList = new Array<string>(helmCol.length);
+    // this function is paralelized and in threads, so will not block the UI. OFC, we prefer to use it.
+    // if not found, we will use the default one running in main thread...
     const beautifyMolsChemFunc = DG.Func.find({package: 'Chem', name: 'beautifyMols'})[0];
-    // both options set to false, ugly molfiles
+    // similarly, OCL Function is also paralelized and in threads, so will not block the UI.
+    const OCLFunc = DG.Func.find({package: 'Chem', name: 'convertToV3KViaOCL'})[0];
+
+    // depending on the function found, we will use it or not. if not, use internal OCL and beautification
     for (let i = 0; i < helmCol.length; i++) {
       molList[i] = (this.helmToAtomicLevelSingle(helmList[i], converter,
-        chiralityEngine, !beautifyMolsChemFunc)).molfile;
+        chiralityEngine && !OCLFunc, !beautifyMolsChemFunc)).molfile;
     }
     // need to beautify the molfiles
     if (beautifyMolsChemFunc) {
@@ -103,6 +108,15 @@ export class SeqHelper implements ISeqHelper {
           beautifiedMols[i] && (molList[i] = beautifiedMols[i]);
       }
     }
+    // handle OCL
+    if (chiralityEngine && OCLFunc) {
+      const oclMols = await OCLFunc.apply({mols: molList});
+      if (oclMols && Array.isArray(oclMols) && oclMols.length === helmCol.length) {
+        for (let i = 0; i < helmCol.length; i++)
+          oclMols[i] && (molList[i] = oclMols[i]);
+      } else
+        grok.shell.warning('OCL function returned an unexpected result');
+    }
     //const molHlList = molfilesV3K.map((item: MolfileWithMap) => getMolHighlight(item.monomers.values(), monomerLib));
 
     const molCol = DG.Column.fromStrings(molColName, molList);