@datagrok/bio 1.11.1 → 1.11.2

package/package.json

@@ -5,7 +5,7 @@
     "name": "Leonid Stolbov",
     "email": "lstolbov@datagrok.ai"
   },
-  "version": "1.11.1",
+  "version": "1.11.2",
   "description": "Bio is a [package](https://datagrok.ai/help/develop/develop#packages) for the [Datagrok](https://datagrok.ai) platform",
   "repository": {
     "type": "git",
@@ -14,12 +14,12 @@
   },
   "dependencies": {
     "@biowasm/aioli": ">=2.4.0",
-    "@datagrok-libraries/bio": "^4.2.0",
+    "@datagrok-libraries/bio": "^4.2.1",
     "@datagrok-libraries/chem-meta": "1.0.0",
     "@datagrok-libraries/ml": "^6.0.0",
     "@datagrok-libraries/utils": "^1.6.2",
     "cash-dom": "latest",
-    "datagrok-api": "^1.6.6",
+    "datagrok-api": "^1.6.7",
     "dayjs": "^1.11.4",
     "openchemlib": "6.0.1",
     "rxjs": "^6.5.5",

package/src/tests/checkInputColumn-tests.ts

@@ -4,7 +4,7 @@ import * as grok from 'datagrok-api/grok';
 import * as ui from 'datagrok-api/ui';
 import * as DG from 'datagrok-api/dg';
 
-import {checkInputColumn} from '../package';
+import {checkInputColumn, multipleSequenceAlignmentAny} from '../package';
 import {UNITS} from 'datagrok-api/dg';
 import {ALPHABET, UnitsHandler} from '@datagrok-libraries/bio/src/utils/units-handler';
 
@@ -17,53 +17,56 @@ seq2,
 seq3,
 seq4`;
 
-  category('MSA', () => {
+  test('testMsaPos', async () => {
+    const func: DG.Func = DG.Func.find({package: 'Bio', name: 'multipleSequenceAlignmentAny'})[0];
+    const funcInputColumnProperty: DG.Property = func.inputs.find((i) => i.name == 'sequence')!;
 
-    test('testMsaPos', async () => {
-      const func: DG.Func = DG.Func.find({package: 'Bio', name: 'multipleSequenceAlignmentAny'})[0];
-      const funcInputColumnProperty: DG.Property = func.inputs.find((i) => i.name == 'sequence')!;
+    let k = 11;
 
-      let k = 11;
+    const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
+    const col: DG.Column = df.getCol('seq');
+    col.semType = DG.SEMTYPE.MACROMOLECULE;
+    col.setTag(DG.TAGS.UNITS, 'fasta');
+    col.setTag(UnitsHandler.TAGS.alphabet, ALPHABET.DNA);
 
-      const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
-      const col: DG.Column = df.getCol('seq');
-      col.semType = DG.SEMTYPE.MACROMOLECULE;
-      col.setTag(DG.TAGS.UNITS, 'fasta');
-      col.setTag(UnitsHandler.TAGS.alphabet, ALPHABET.DNA);
+    const [res, msg]: [boolean, string] = checkInputColumn(
+      col, 'Test', ['fasta',], ['DNA', 'RNA', 'PT']);
 
-      const [res, msg]: [boolean, string] = checkInputColumn(
-        col, 'Test', ['fasta',], ['DNA', 'RNA', 'PT']);
+    expect(res, true);
+  });
 
-      expect(res, true);
-    });
+  test('testMsaNegHelm', async () => {
+    const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
+    const col: DG.Column = df.getCol('seq');
+    col.semType = DG.SEMTYPE.MACROMOLECULE;
+    col.setTag(DG.TAGS.UNITS, 'helm');
+    col.setTag(UnitsHandler.TAGS.alphabetSize, '11');
+    col.setTag(UnitsHandler.TAGS.alphabetIsMultichar, 'true');
 
-    test('testMsaNegHelm', async () => {
-      const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
-      const col: DG.Column = df.getCol('seq');
-      col.semType = DG.SEMTYPE.MACROMOLECULE;
-      col.setTag(DG.TAGS.UNITS, 'helm');
-      col.setTag(UnitsHandler.TAGS.alphabetSize, '11');
-      col.setTag(UnitsHandler.TAGS.alphabetIsMultichar, 'true');
+    const [res, msg]: [boolean, string] = checkInputColumn(
+      col, 'Test', ['fasta',], ['DNA', 'RNA', 'PT']);
 
-      const [res, msg]: [boolean, string] = checkInputColumn(
-        col, 'Test', ['fasta',], ['DNA', 'RNA', 'PT']);
+    expect(res, false);
+  });
 
-      expect(res, false);
-    });
+  test('testMsaNegUN', async () => {
+    const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
+    const col: DG.Column = df.getCol('seq');
+    col.semType = DG.SEMTYPE.MACROMOLECULE;
+    col.setTag(DG.TAGS.UNITS, 'fasta');
+    col.setTag(UnitsHandler.TAGS.alphabet, 'UN');
+    col.setTag(UnitsHandler.TAGS.alphabetSize, '11');
+    col.setTag(UnitsHandler.TAGS.alphabetIsMultichar, 'true');
 
-    test('testMsaNegUN', async () => {
-      const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
-      const col: DG.Column = df.getCol('seq');
-      col.semType = DG.SEMTYPE.MACROMOLECULE;
-      col.setTag(DG.TAGS.UNITS, 'fasta');
-      col.setTag(UnitsHandler.TAGS.alphabet, 'UN');
-      col.setTag(UnitsHandler.TAGS.alphabetSize, '11');
-      col.setTag(UnitsHandler.TAGS.alphabetIsMultichar, 'true');
+    const [res, msg]: [boolean, string] = checkInputColumn(
+      col, 'Test', ['fasta',], ['DNA', 'RNA', 'PT']);
 
-      const [res, msg]: [boolean, string] = checkInputColumn(
-        col, 'Test', ['fasta',], ['DNA', 'RNA', 'PT']);
+    expect(res, false);
+  });
 
-      expect(res, false);
-    });
+  test('testGetActionFunctionMeta', async () => {
+    const func: DG.Func = DG.Func.find({package: 'Bio', name: 'multipleSequenceAlignmentAny'})[0];
+    const sequenceInput: DG.Property = func.inputs.find((i) => i.name == 'sequence')!;
+    let k = 11;
   });
 });

package/src/tests/convert-test.ts

@@ -5,7 +5,7 @@ import * as DG from 'datagrok-api/dg';
 
 import {ConverterFunc} from './types';
 import {NotationConverter} from '@datagrok-libraries/bio/src/utils/notation-converter';
-import {NOTATION} from '@datagrok-libraries/bio/src/utils/units-handler';
+import {NOTATION, UnitsHandler} from '@datagrok-libraries/bio/src/utils/units-handler';
 
 // import {mmSemType} from '../const';
 // import {importFasta} from '../package';
@@ -145,14 +145,17 @@ RNA1{P.R(U)P.R(U)P.R(C)P.R(A)P.R(A)P.R(C)P.P.P}$$$
 
   async function _testConvert(srcKey: string, converter: ConverterFunc, tgtKey: string) {
     const srcDf: DG.DataFrame = await readCsv(srcKey);
-    const srcCol: DG.Column = srcDf.col('seq')!;
+    const srcCol: DG.Column = srcDf.getCol('seq');
 
+    // conversion results
     const resCol: DG.Column = converter(srcCol);
 
+    // The correct reference data to compare conversion results with.
     const tgtDf: DG.DataFrame = await readCsv(tgtKey);
-    const tgtCol: DG.Column = tgtDf.col('seq')!;
+    const tgtCol: DG.Column = tgtDf.getCol('seq');
 
     expectArray(resCol.toList(), tgtCol.toList());
+    const uh: UnitsHandler = new UnitsHandler(resCol);
   }
 
   // FASTA tests

package/src/tests/renderers-test.ts

@@ -3,10 +3,9 @@ import {after, before, category, delay, expect, test} from '@datagrok-libraries/
 import * as grok from 'datagrok-api/grok';
 import * as DG from 'datagrok-api/dg';
 import {importFasta, multipleSequenceAlignmentAny} from '../package';
-import {readDataframe} from './utils';
 import {convertDo} from '../utils/convert';
 import {ALPHABET, NOTATION, UnitsHandler} from '@datagrok-libraries/bio/src/utils/units-handler';
-import { SEM_TYPES, TAGS } from '../utils/constants';
+import {SEM_TYPES, TAGS} from '../utils/constants';
 
 category('renderers', () => {
   let tvList: DG.TableView[];
@@ -38,30 +37,38 @@ category('renderers', () => {
   async function _testAfterMsa() {
     const fastaTxt: string = await grok.dapi.files.readAsText('System:AppData/Bio/samples/sample_FASTA.fasta');
     const df: DG.DataFrame = importFasta(fastaTxt)[0];
+
+    const srcSeqCol: DG.Column = df.getCol('sequence');
+    const semType: string = await grok.functions.call('Bio:detectMacromolecule', {col: srcSeqCol});
+    if (semType)
+      srcSeqCol.semType = semType;
+
     const tv: DG.TableView = grok.shell.addTableView(df);
+    // call to calculate 'cell.renderer' tag
     await grok.data.detectSemanticTypes(df);
-    console.log('Bio: tests/renderers/afterMsa, table view');
 
-    const srcSeqCol: DG.Column | null = df.col('sequence');
-    expect(srcSeqCol !== null, true);
+    console.log('Bio: tests/renderers/afterMsa, table view');
 
     console.log('Bio: tests/renderers/afterMsa, src before test ' +
       `semType="${srcSeqCol!.semType}", units="${srcSeqCol!.getTag(DG.TAGS.UNITS)}", ` +
       `cell.renderer="${srcSeqCol!.getTag(DG.TAGS.CELL_RENDERER)}"`);
-    expect(srcSeqCol!.semType, DG.SEMTYPE.MACROMOLECULE);
-    expect(srcSeqCol!.getTag(DG.TAGS.UNITS), NOTATION.FASTA);
-    expect(srcSeqCol!.getTag(UnitsHandler.TAGS.aligned), 'SEQ');
-    expect(srcSeqCol!.getTag(UnitsHandler.TAGS.alphabet), ALPHABET.PT);
-    expect(srcSeqCol!.getTag(DG.TAGS.CELL_RENDERER), 'sequence');
+    expect(srcSeqCol.semType, DG.SEMTYPE.MACROMOLECULE);
+    expect(srcSeqCol.getTag(DG.TAGS.UNITS), NOTATION.FASTA);
+    expect(srcSeqCol.getTag(UnitsHandler.TAGS.aligned), 'SEQ');
+    expect(srcSeqCol.getTag(UnitsHandler.TAGS.alphabet), ALPHABET.PT);
+    expect(srcSeqCol.getTag(DG.TAGS.CELL_RENDERER), 'sequence');
 
-    const msaSeqCol: DG.Column | null = await multipleSequenceAlignmentAny(df, srcSeqCol!);
+    const msaSeqCol: DG.Column = (await multipleSequenceAlignmentAny(df, srcSeqCol!))!;
     tv.grid.invalidate();
 
-    expect(msaSeqCol!.semType, DG.SEMTYPE.MACROMOLECULE);
-    expect(msaSeqCol!.getTag(DG.TAGS.UNITS), NOTATION.FASTA);
-    expect(msaSeqCol!.getTag(UnitsHandler.TAGS.aligned), 'SEQ.MSA');
-    expect(msaSeqCol!.getTag(UnitsHandler.TAGS.alphabet), ALPHABET.PT);
-    expect(msaSeqCol!.getTag(DG.TAGS.CELL_RENDERER), 'sequence');
+    expect(msaSeqCol.semType, DG.SEMTYPE.MACROMOLECULE);
+    expect(msaSeqCol.getTag(DG.TAGS.UNITS), NOTATION.FASTA);
+    expect(msaSeqCol.getTag(UnitsHandler.TAGS.aligned), 'SEQ.MSA');
+    expect(msaSeqCol.getTag(UnitsHandler.TAGS.alphabet), ALPHABET.PT);
+    expect(msaSeqCol.getTag(DG.TAGS.CELL_RENDERER), 'sequence');
+
+    // check newColumn with UnitsHandler constructor
+    const uh: UnitsHandler = new UnitsHandler(msaSeqCol);
 
     dfList.push(df);
     tvList.push(tv);
@@ -71,25 +78,35 @@ category('renderers', () => {
     const csv: string = await grok.dapi.files.readAsText('System:AppData/Bio/samples/sample_FASTA_PT.csv');
     const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
     const tv: DG.TableView = grok.shell.addTableView(df);
-    await grok.data.detectSemanticTypes(df);
 
     const srcCol: DG.Column = df.col('sequence')!;
+    // await grok.data.detectSemanticTypes(df);
+    const semType: string = await grok.functions.call('Bio:detectMacromolecule', {col: srcCol});
+    if (semType)
+      srcCol.semType = semType;
+    await grok.data.detectSemanticTypes(df);
+
     const tgtCol: DG.Column = await convertDo(srcCol, NOTATION.SEPARATOR, '/');
     expect(tgtCol.getTag(DG.TAGS.CELL_RENDERER), 'sequence');
 
+    // check tgtCol with UnitsHandler constructor
+    const uh: UnitsHandler = new UnitsHandler(tgtCol);
+
     tvList.push(tv);
     dfList.push(df);
-  };
+  }
 
   async function _setRendererManually() {
-    const df = DG.DataFrame.fromColumns([DG.Column.fromStrings('SequencesDiff', ['meI/hHis/Aca/N/T/dK/Thr_PO3H2/Aca#D-Tyr_Et/Tyr_ab-dehydroMe/meN/E/N/dV'])]);
+    const df = DG.DataFrame.fromColumns([DG.Column.fromStrings(
+      'SequencesDiff', ['meI/hHis/Aca/N/T/dK/Thr_PO3H2/Aca#D-Tyr_Et/Tyr_ab-dehydroMe/meN/E/N/dV'])]);
     df.col('SequencesDiff')!.tags[DG.TAGS.UNITS] = 'separator';
     df.col('SequencesDiff')!.tags[TAGS.SEPARATOR] = '/';
     df.col('SequencesDiff')!.semType = SEM_TYPES.MACROMOLECULE_DIFFERENCE;
     const tw = grok.shell.addTableView(df);
     await delay(100);
     const renderer = tw.dataFrame.col('SequencesDiff')?.getTag(DG.TAGS.CELL_RENDERER);
-    if (renderer !== 'MacromoleculeDifferenceCR') 
-      throw new Error(`Units 'separator', separator '/' and semType 'MacromoleculeDifference' have been manually set on column but after df aws added as table view renderer has been reset to '${renderer}'`)
-  };
+    if (renderer !== 'MacromoleculeDifferenceCR')
+      throw new Error(`Units 'separator', separator '/' and semType 'MacromoleculeDifference' have been ` +
+        `manually set on column but after df aws added as table view renderer has been reset to '${renderer}'`);
+  }
 });

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

@@ -1,10 +1,15 @@
 import * as DG from 'datagrok-api/dg';
 import * as grok from 'datagrok-api/grok';
-import { expect } from '@datagrok-libraries/utils/src/test';
-import { sequenceSpaceTopMenu } from '../package';
+import {expect} from '@datagrok-libraries/utils/src/test';
+import {sequenceSpaceTopMenu} from '../package';
 
 export async function _testSequenceSpaceReturnsResult(df: DG.DataFrame, algorithm: string, colName: string) {
-  await grok.data.detectSemanticTypes(df);
+  // await grok.data.detectSemanticTypes(df);
+  const col: DG.Column = df.getCol(colName);
+  const semType: string = await grok.functions.call('Bio:detectMacromolecule', {col: col});
+  if (semType)
+    col.semType = semType;
+
   const sp = await sequenceSpaceTopMenu(df, df.col(colName)!, algorithm, 'Levenshtein', true);
   expect(sp != null, true);
 }

package/src/utils/convert.ts

@@ -5,7 +5,7 @@ import $ from 'cash-dom';
 
 import {Subscription} from 'rxjs';
 import {NotationConverter} from '@datagrok-libraries/bio/src/utils/notation-converter';
-import {NOTATION} from '@datagrok-libraries/bio/src/utils/units-handler';
+import {NOTATION, UnitsHandler} from '@datagrok-libraries/bio/src/utils/units-handler';
 
 
 let convertDialog: DG.Dialog | null = null;
@@ -85,6 +85,9 @@ export async function convertDo(
   const converter = new NotationConverter(srcCol);
   const newColumn = converter.convert(targetNotation, separator);
   srcCol.dataFrame.columns.add(newColumn);
+
+  // call to calculate 'cell.renderer' tag
   await grok.data.detectSemanticTypes(srcCol.dataFrame);
+
   return newColumn;
 }

package/{test-Bio-f1ac5a5eade4-e2085bf5.html → test-Bio-f1ac5a5eade4-a0f7e8c0.html}

@@ -1,4 +1,4 @@
-<html><head><meta charset="utf-8"/><title>Bio Test Report. Datagrok version datagrok/datagrok:latest SHA=f1ac5a5eade4. Commit e2085bf5.</title><style type="text/css">html,
+<html><head><meta charset="utf-8"/><title>Bio Test Report. Datagrok version datagrok/datagrok:latest SHA=f1ac5a5eade4. Commit a0f7e8c0.</title><style type="text/css">html,
 body {
   font-family: Arial, Helvetica, sans-serif;
   font-size: 1rem;
@@ -229,7 +229,7 @@ header {
   font-size: 1rem;
   padding: 0 0.5rem;
 }
-</style></head><body><div id="jesthtml-content"><header><h1 id="title">Bio Test Report. Datagrok version datagrok/datagrok:latest SHA=f1ac5a5eade4. Commit e2085bf5.</h1></header><div id="metadata-container"><div id="timestamp">Started: 2022-09-08 09:07:23</div><div id="summary"><div id="suite-summary"><div class="summary-total">Suites (1)</div><div class="summary-passed summary-empty">0 passed</div><div class="summary-failed">1 failed</div><div class="summary-pending summary-empty">0 pending</div></div><div id="test-summary"><div class="summary-total">Tests (1)</div><div class="summary-passed summary-empty">0 passed</div><div class="summary-failed">1 failed</div><div class="summary-pending summary-empty">0 pending</div></div></div></div><div id="suite-1" class="suite-container"><div class="suite-info"><div class="suite-path">/home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts</div><div class="suite-time warn">10.876s</div></div><div class="suite-tests"><div class="test-result failed"><div class="test-info"><div class="test-suitename"> </div><div class="test-title">TEST</div><div class="test-status">failed</div><div class="test-duration">0s</div></div><div class="failureMessages"> <pre class="failureMsg">Error: Unknown server alias. Please add it to /home/runner/.grok/config.yaml
+</style></head><body><div id="jesthtml-content"><header><h1 id="title">Bio Test Report. Datagrok version datagrok/datagrok:latest SHA=f1ac5a5eade4. Commit a0f7e8c0.</h1></header><div id="metadata-container"><div id="timestamp">Started: 2022-09-08 10:03:50</div><div id="summary"><div id="suite-summary"><div class="summary-total">Suites (1)</div><div class="summary-passed summary-empty">0 passed</div><div class="summary-failed">1 failed</div><div class="summary-pending summary-empty">0 pending</div></div><div id="test-summary"><div class="summary-total">Tests (1)</div><div class="summary-passed summary-empty">0 passed</div><div class="summary-failed">1 failed</div><div class="summary-pending summary-empty">0 pending</div></div></div></div><div id="suite-1" class="suite-container"><div class="suite-info"><div class="suite-path">/home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts</div><div class="suite-time warn">10.76s</div></div><div class="suite-tests"><div class="test-result failed"><div class="test-info"><div class="test-suitename"> </div><div class="test-title">TEST</div><div class="test-status">failed</div><div class="test-duration">0s</div></div><div class="failureMessages"> <pre class="failureMsg">Error: Unknown server alias. Please add it to /home/runner/.grok/config.yaml
     at getDevKey (/home/runner/work/public/public/packages/Bio/src/__jest__/test-node.ts:48:13)
     at Object.&lt;anonymous&gt; (/home/runner/work/public/public/packages/Bio/src/__jest__/test-node.ts:57:15)
     at Generator.next (&lt;anonymous&gt;)