@datagrok/bio 1.5.9 → 1.5.10

package/package.json

@@ -2,7 +2,7 @@
   "name": "@datagrok/bio",
   "beta": false,
   "friendlyName": "Bio",
-  "version": "1.5.9",
+  "version": "1.5.10",
   "description": "Bio is a [package](https://datagrok.ai/help/develop/develop#packages) for the [Datagrok](https://datagrok.ai) platform",
   "repository": {
     "type": "git",
@@ -11,7 +11,7 @@
   },
   "dependencies": {
     "@biowasm/aioli": ">=2.4.0",
-    "@datagrok-libraries/bio": "^2.4.1",
+    "@datagrok-libraries/bio": "^2.4.2",
     "@datagrok-libraries/utils": "^1.0.0",
     "@datagrok-libraries/ml": "^2.0.9",
     "cash-dom": "latest",

package/src/package.ts

@@ -135,6 +135,11 @@ export async function sequenceSpaceTopMenu(table: DG.DataFrame, macroMolecule: D
 //input: dataframe df [Input data table]
 //input: column sequence {semType: Macromolecule}
 export async function toAtomicLevel(df: DG.DataFrame, sequence: DG.Column): Promise<void> {
+  if (DG.Func.find({package: 'Chem', name: 'getRdKitModule'})) {
+    grok.shell.warning('Transformation to atomic level requires package "Chem" installed.');
+    return;
+  }
+
   const monomersLibFile = await _package.files.readAsText(HELM_CORE_LIB_FILENAME);
   const monomersLibObject: any[] = JSON.parse(monomersLibFile);
   const atomicCodes = getMolfilesFromSeq(sequence, monomersLibObject);
@@ -152,7 +157,23 @@ export async function toAtomicLevel(df: DG.DataFrame, sequence: DG.Column): Prom
 //input: dataframe table
 //input: column sequence { semType: Macromolecule }
 //output: column result
-export async function multipleSequenceAlignmentAny(table: DG.DataFrame, col: DG.Column): Promise<DG.Column> {
+export async function multipleSequenceAlignmentAny(table: DG.DataFrame, col: DG.Column): Promise<DG.Column | null> {
+  if (col.semType != DG.SEMTYPE.MACROMOLECULE) {
+    grok.shell.warning(`MSA analysis is allowed for semantic type "${DG.SEMTYPE.MACROMOLECULE}" data only.`);
+    return null;
+  }
+  const units: string = col.getTag(DG.TAGS.UNITS);
+  const allowedAlphabets = ['DNA', 'RNA', 'PT'];
+  const allowedNotations = ['fasta'];
+  if (!allowedAlphabets.some((a) => units.toUpperCase().endsWith(a.toUpperCase())) ||
+    !allowedNotations.some((n) => units.toUpperCase().startsWith(n.toUpperCase()))) {
+    grok.shell.warning('MSA analysis is allowed for ' +
+      `notation${allowedNotations.length > 1 ? 's' : ''} ${allowedNotations.map((n) => `"${n}"`).join(', ')} ` +
+      'and ' +
+      `alphabet${allowedAlphabets.length > 1 ? 's' : ''} ${allowedAlphabets.map((a) => `"${a}"`).join(', ')}.`);
+    return null;
+  }
+
   const msaCol = await runKalign(col, false);
   table.columns.add(msaCol);
 
@@ -171,19 +192,21 @@ export async function compositionAnalysis(): Promise<void> {
   // Higher priority for columns with MSA data to show with WebLogo.
   const tv = grok.shell.tv;
   const df = tv.dataFrame;
-  const semTypeColList = df.columns.bySemTypeAll(DG.SEMTYPE.MACROMOLECULE);
-  let col: DG.Column | undefined = semTypeColList.find((col) => {
-    const units = col.getTag(DG.TAGS.UNITS);
-    return units ? units.indexOf('MSA') !== -1 : false;
-  });
-  if (!col)
-    col = semTypeColList[0];
 
+  const col: DG.Column | null = WebLogo.pickUpSeqCol2(df);
   if (!col) {
     grok.shell.error('Current table does not contain sequences');
     return;
   }
 
+  const allowedNotations: string[] = ['fasta', 'separator'];
+  const units = col.getTag(DG.TAGS.UNITS);
+  if (!allowedNotations.some((n) => units.toUpperCase().startsWith(n.toUpperCase()))) {
+    grok.shell.warning('Composition analysis is allowed for ' +
+      `notation${allowedNotations.length > 1 ? 's' : ''} ${allowedNotations.map((n) => `"${n}"`).join(', ')}.`);
+    return;
+  }
+
   tv.addViewer('WebLogo', {sequenceColumnName: col.name});
 }
 

package/src/tests/detectors-test.ts

@@ -106,10 +106,11 @@ MWRSWY-CKHP
     fastaFasta = 'fastaFasta',
     msaComplex = 'msaComplex',
     helmCsv = 'helmCsv',
+    testDemogCsv = 'testDemogCsv',
+    testHelmCsv = 'testHelmCsv',
     testIdCsv = 'testIdCsv',
     testSmilesCsv = 'testSmilesCsv',
-    testHelmCsv = 'testHelmCsv',
-    testDemogCsv = 'testDemogCsv',
+    testSmiles2Csv = 'testSmiles2Csv',
   }
 
   const samples: { [key: string]: string } = {
@@ -119,9 +120,10 @@ MWRSWY-CKHP
     'msaComplex': 'System:AppData/Bio/samples/sample_MSA.csv',
     'helmCsv': 'System:AppData/Bio/samples/sample_HELM.csv',
     'testDemogCsv': 'System:AppData/Bio/samples/testDemog.csv',
-    'testIdCsv': 'System:AppData/Bio/samples/id.csv',
     'testHelmCsv': 'System:AppData/Bio/samples/testHelm.csv',
+    'testIdCsv': 'System:AppData/Bio/samples/id.csv',
     'testSmilesCsv': 'System:AppData/Bio/samples/testSmiles.csv',
+    'testSmiles2Csv': 'System:AppData/Bio/samples/testSmiles2.csv',
   };
 
   const _samplesDfs: { [key: string]: Promise<DG.DataFrame> } = {};
@@ -288,6 +290,10 @@ MWRSWY-CKHP
     for (const col of df.columns.toList())
       await _testNeg(dfFunc, col.name);
   });
+
+  test('samplesTestSmiles2NegativeSmiles', async () => {
+    await _testNeg(readSamples(Samples.testSmiles2Csv), 'SMILES');
+  });
 });
 
 export async function _testNeg(readDf: DfReaderFunc, colName: string) {

package/src/tests/renderer-test.ts

@@ -6,31 +6,53 @@ import * as DG from 'datagrok-api/dg';
 import {importFasta, multipleSequenceAlignmentAny} from '../package';
 
 category('renderers', () => {
+  let tvList: DG.TableView[];
+
+  before(async () => {
+    tvList = [];
+  });
+
+  after(async () => {
+    tvList.forEach((tv: DG.TableView) => tv.close());
+  });
+
   test('afterMsa', async () => {
     await _testAfterMsa();
   });
-});
 
-export 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];
+  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 | null = df.col('sequence');
+    expect(srcSeqCol !== null, true);
+    console.log('Bio: tests/renderers/afterMsa, src data loaded');
 
-  const seqCol: DG.Column | null = df.col('sequence');
-  expect(seqCol !== null, true);
+    const tv: DG.TableView = grok.shell.addTableView(df);
+    console.log('Bio: tests/renderers/afterMsa, table view');
 
-  const tv: DG.TableView = grok.shell.addTableView(df);
-  await grok.data.detectSemanticTypes(df);
+    await grok.data.detectSemanticTypes(df);
+    console.log('Bio: tests/renderers/afterMsa, detectSemanticTypes');
 
-  expect(seqCol!.semType, DG.SEMTYPE.MACROMOLECULE);
-  expect(seqCol!.getTag(DG.TAGS.UNITS), 'fasta:SEQ:PT');
-  expect(seqCol!.getTag('cell.renderer'), 'Macromolecule');
+    console.log('Bio: tests/renderers/afterMsa, src before test semType' +
+      `semType="${srcSeqCol!.semType}", units="${srcSeqCol!.getTag(DG.TAGS.UNITS)}", ` +
+      `cell.renderer="${srcSeqCol!.getTag('cell.renderer')}"`);
+    expect(srcSeqCol!.semType, DG.SEMTYPE.MACROMOLECULE);
+    expect(srcSeqCol!.getTag(DG.TAGS.UNITS), 'fasta:SEQ:PT');
+    expect(srcSeqCol!.getTag('cell.renderer'), 'Macromolecule');
+    console.log('Bio: tests/renderers/afterMsa, src semType tested');
 
-  const seqMsaCol: DG.Column = await multipleSequenceAlignmentAny(df, seqCol!);
-  tv.grid.invalidate();
+    const msaSeqCol: DG.Column | null = await multipleSequenceAlignmentAny(df, srcSeqCol!);
+    console.log('Bio: tests/renderers/afterMsa, msaSeqCol created');
 
-  expect(seqMsaCol!.semType, DG.SEMTYPE.MACROMOLECULE);
-  expect(seqMsaCol!.getTag(DG.TAGS.UNITS), 'fasta:SEQ.MSA:PT');
-  expect(seqMsaCol!.getTag('cell.renderer'), 'Macromolecule');
+    tv.grid.invalidate();
+    console.log('Bio: tests/renderers/afterMsa, tv.grid invalidated');
 
-  // tv.close();
-}
+    expect(msaSeqCol!.semType, DG.SEMTYPE.MACROMOLECULE);
+    expect(msaSeqCol!.getTag(DG.TAGS.UNITS), 'fasta:SEQ.MSA:PT');
+    expect(msaSeqCol!.getTag('cell.renderer'), 'Macromolecule');
+    console.log('Bio: tests/renderers/afterMsa, msa semType tested');
+
+    tvList.push(tv);
+  }
+});

package/src/utils/cell-renderer.ts

@@ -193,7 +193,7 @@ export class MacromoleculeSequenceCellRenderer extends DG.GridCellRenderer {
       const splitterFunc: SplitterFunc = WebLogo.getSplitter(units, gridCell.cell.column.getTag('separator') );// splitter,
 
       const subParts:string[] =  splitterFunc(cell.value);
-      console.log(subParts);
+      // console.log(subParts);
 
       const textSize = g.measureText(subParts.join(''));
       let x1 = Math.max(x, x + (w - textSize.width) / 2);

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

@@ -50,10 +50,12 @@ export async function runKalign(srcCol: DG.Column, isAligned = false): Promise<D
 
   await CLI.fs.writeFile('input.fa', fasta);
   const output = await CLI.exec('kalign input.fa -f fasta -o result.fasta');
-  const buf = await CLI.cat('result.fasta');
-
   console.warn(output);
 
+  const buf = await CLI.cat('result.fasta');
+  if (!buf)
+    throw new Error(`kalign output no result`);
+
   const aligned = _fastaToStrings(buf).slice(0, sequences.length);
   const tgtCol = DG.Column.fromStrings(`msa(${srcCol.name})`, aligned);
 

package/{test-Bio-34f75e5127b8-b47d4664.html → test-Bio-34f75e5127b8-936bf89b.html}

@@ -1,4 +1,4 @@
-<html><head><meta charset="utf-8"/><title>Bio Test Report. Datagrok version datagrok/datagrok:latest SHA=34f75e5127b8. Commit b47d4664.</title><style type="text/css">html,
+<html><head><meta charset="utf-8"/><title>Bio Test Report. Datagrok version datagrok/datagrok:latest SHA=34f75e5127b8. Commit 936bf89b.</title><style type="text/css">html,
 body {
   font-family: Arial, Helvetica, sans-serif;
   font-size: 1rem;
@@ -229,14 +229,21 @@ 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=34f75e5127b8. Commit b47d4664.</h1></header><div id="metadata-container"><div id="timestamp">Started: 2022-07-11 12:49:26</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">107.855s</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">98.311s</div></div><div class="failureMessages"> <pre class="failureMsg">Error: expect(received).toBe(expected) // Object.is equality
-
-Expected: false
-Received: true
-    at /home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:49:20
-    at Generator.next (&lt;anonymous&gt;)
-    at fulfilled (/home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:31:58)
-    at processTicksAndRejections (internal/process/task_queues.js:97:5)</pre></div></div></div><div class="suite-consolelog"><div class="suite-consolelog-header">Console Log</div><div class="suite-consolelog-item"><pre class="suite-consolelog-item-origin">    at Object.&lt;anonymous&gt; (/home/runner/work/public/public/packages/Bio/src/__jest__/test-node.ts:63:11)
+</style></head><body><div id="jesthtml-content"><header><h1 id="title">Bio Test Report. Datagrok version datagrok/datagrok:latest SHA=34f75e5127b8. Commit 936bf89b.</h1></header><div id="metadata-container"><div id="timestamp">Started: 2022-07-11 18:26:55</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">109.382s</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">100.002s</div></div><div class="failureMessages"> <pre class="failureMsg">Error: thrown: "Exceeded timeout of 100000 ms for a test.
+Use jest.setTimeout(newTimeout) to increase the timeout value, if this is a long-running test."
+    at Object.&lt;anonymous&gt; (/home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:22:1)
+    at Runtime._execModule (/home/runner/work/public/public/packages/Bio/node_modules/jest-runtime/build/index.js:1646:24)
+    at Runtime._loadModule (/home/runner/work/public/public/packages/Bio/node_modules/jest-runtime/build/index.js:1185:12)
+    at Runtime.requireModule (/home/runner/work/public/public/packages/Bio/node_modules/jest-runtime/build/index.js:1009:12)
+    at jestAdapter (/home/runner/work/public/public/packages/Bio/node_modules/jest-circus/build/legacy-code-todo-rewrite/jestAdapter.js:79:13)
+    at runTestInternal (/home/runner/work/public/public/packages/Bio/node_modules/jest-runner/build/runTest.js:389:16)
+    at runTest (/home/runner/work/public/public/packages/Bio/node_modules/jest-runner/build/runTest.js:475:34)
+    at TestRunner.runTests (/home/runner/work/public/public/packages/Bio/node_modules/jest-runner/build/index.js:101:12)
+    at TestScheduler.scheduleTests (/home/runner/work/public/public/packages/Bio/node_modules/@jest/core/build/TestScheduler.js:333:13)
+    at runJest (/home/runner/work/public/public/packages/Bio/node_modules/@jest/core/build/runJest.js:404:19)
+    at _run10000 (/home/runner/work/public/public/packages/Bio/node_modules/@jest/core/build/cli/index.js:320:7)
+    at runCLI (/home/runner/work/public/public/packages/Bio/node_modules/@jest/core/build/cli/index.js:173:3)
+    at Object.run (/home/runner/work/public/public/packages/Bio/node_modules/jest-cli/build/cli/index.js:155:37)</pre></div></div></div><div class="suite-consolelog"><div class="suite-consolelog-header">Console Log</div><div class="suite-consolelog-item"><pre class="suite-consolelog-item-origin">    at Object.&lt;anonymous&gt; (/home/runner/work/public/public/packages/Bio/src/__jest__/test-node.ts:63:11)
     at Generator.next (&lt;anonymous&gt;)
     at fulfilled (/home/runner/work/public/public/packages/Bio/src/__jest__/test-node.ts:28:58)
     at processTicksAndRejections (internal/process/task_queues.js:97:5)</pre><pre class="suite-consolelog-item-message">Using web root: http://localhost:8080</pre></div><div class="suite-consolelog-item"><pre class="suite-consolelog-item-origin">    at /home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:24:11
@@ -246,15 +253,4 @@ Received: true
     at Object.&lt;anonymous&gt;.__awaiter (/home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:30:12)
     at Object.&lt;anonymous&gt; (/home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:22:23)
     at Promise.then.completed (/home/runner/work/public/public/packages/Bio/node_modules/jest-circus/build/utils.js:391:28)
-    at new Promise (&lt;anonymous&gt;)</pre><pre class="suite-consolelog-item-message">Testing Bio package</pre></div><div class="suite-consolelog-item"><pre class="suite-consolelog-item-origin">    at /home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:47:11
-    at Generator.next (&lt;anonymous&gt;)
-    at fulfilled (/home/runner/work/public/public/packages/Bio/src/__jest__/remote.test.ts:31:58)
-    at processTicksAndRejections (internal/process/task_queues.js:97:5)</pre><pre class="suite-consolelog-item-message">detectors.samplesIdCsvNegativeID: Operation caused an exception (FileSystemException: Cannot open file, path = '/home/grok/data/prod/packages/data/Bio/samples/id.csv' (OS Error: No such file or directory, errno = 2))
-MSA.is_correct: TypeError: Cannot read properties of undefined (reading 'split')
-sequenceSpace.sequenceSpaceOpens: TypeError: Cannot read properties of undefined (reading 'col')
-sequenceSpace.init: Operation caused an exception (FileSystemException: Cannot open file, path = '/home/grok/data/prod/packages/data/Bio/sample_FASTA.csv' (OS Error: No such file or directory, errno = 2))
-activityCliffs.activityCliffsOpen: TypeError: Cannot read properties of undefined (reading 'columns')
-activityCliffs.init: TypeError: Cannot read properties of undefined (reading 'close')
-activityCliffs.init: Operation caused an exception (FileSystemException: Cannot open file, path = '/home/grok/data/prod/packages/data/Bio/sample_MSA.csv' (OS Error: No such file or directory, errno = 2))
-renderers.afterMsa: Error: Expected "Macromolecule", got "null"
-</pre></div></div></div></div></body></html>
+    at new Promise (&lt;anonymous&gt;)</pre><pre class="suite-consolelog-item-message">Testing Bio package</pre></div></div></div></div></body></html>