@datagrok/bio 1.4.2 → 1.5.3

package/files/samples/id.csv

@@ -0,0 +1,313 @@
+ID
+ID000000670
+ID000000691
+ID000001221
+ID000001243
+ID000001350
+ID000001393
+ID000001466
+ID000001471
+ID000001472
+ID000001590
+ID000001591
+ID000001599
+ID000002433
+ID000002434
+ID000002437
+ID000002440
+ID000002441
+ID000002442
+ID000002443
+ID000002444
+ID000002516
+ID000002878
+ID000002880
+ID000002957
+ID000002998
+ID000003026
+ID000003046
+ID000003200
+ID000003673
+ID000004018
+ID000004090
+ID000004163
+ID000004176
+ID000004222
+ID000004240
+ID000004245
+ID000004730
+ID000004829
+ID000004912
+ID000004969
+ID000005051
+ID000005057
+ID000005060
+ID000005073
+ID000005104
+ID000006942
+ID000007364
+ID000007419
+ID000007757
+ID000007935
+ID000007936
+ID000007946
+ID000008044
+ID000008059
+ID000008118
+ID000010343
+ID000011060
+ID000011866
+ID000011867
+ID000011868
+ID000011869
+ID000011870
+ID000011871
+ID000011927
+ID000011928
+ID000011929
+ID000011930
+ID000011931
+ID000011932
+ID000011933
+ID000011934
+ID000011935
+ID000011943
+ID000011945
+ID000012738
+ID000012739
+ID000012845
+ID000012906
+ID000012932
+ID000012948
+ID000012983
+ID000012989
+ID000012993
+ID000012998
+ID000013000
+ID000013006
+ID000013008
+ID000013034
+ID000013043
+ID000013045
+ID000013046
+ID000013047
+ID000013052
+ID000013054
+ID000013055
+ID000013172
+ID000013174
+ID000013176
+ID000013184
+ID000013199
+ID000013200
+ID000013210
+ID000013215
+ID000013237
+ID000013265
+ID000013270
+ID000013273
+ID000013290
+ID000013332
+ID000013450
+ID000013453
+ID000013492
+ID000013508
+ID000013510
+ID000013588
+ID000013621
+ID000013645
+ID000013658
+ID000013675
+ID000013759
+ID000013769
+ID000013783
+ID000013802
+ID000013822
+ID000013846
+ID000013895
+ID000013896
+ID000013996
+ID000014680
+ID000014681
+ID000014715
+ID000014815
+ID000014870
+ID000015313
+ID000015441
+ID000015447
+ID000015450
+ID000015452
+ID000015453
+ID000015484
+ID000015485
+ID000015489
+ID000015490
+ID000015492
+ID000015493
+ID000015543
+ID000015544
+ID000015618
+ID000015619
+ID000015626
+ID000015649
+ID000015650
+ID000015662
+ID000015669
+ID000015674
+ID000015675
+ID000015676
+ID000015689
+ID000015690
+ID000015695
+ID000015727
+ID000015731
+ID000015739
+ID000015740
+ID000015751
+ID000015753
+ID000015754
+ID000015755
+ID000015758
+ID000015771
+ID000015772
+ID000015773
+ID000015774
+ID000015809
+ID000015829
+ID000015830
+ID000015840
+ID000015841
+ID000015842
+ID000015843
+ID000015845
+ID000015846
+ID000015847
+ID000015848
+ID000015849
+ID000015850
+ID000015851
+ID000015852
+ID000015853
+ID000015864
+ID000015880
+ID000015914
+ID000015940
+ID000015942
+ID000015945
+ID000015956
+ID000015963
+ID000015964
+ID000015968
+ID000015970
+ID000015984
+ID000015985
+ID000016053
+ID000016873
+ID000016946
+ID000017085
+ID000017086
+ID000017164
+ID000017265
+ID000017311
+ID000017366
+ID000017405
+ID000017418
+ID000017494
+ID000017552
+ID000017581
+ID000017589
+ID000017601
+ID000017606
+ID000017607
+ID000017638
+ID000017672
+ID000017774
+ID000017798
+ID000017802
+ID000017822
+ID000017823
+ID000017848
+ID000017880
+ID000017881
+ID000017901
+ID000017990
+ID000018043
+ID000018216
+ID000018619
+ID000018628
+ID000018691
+ID000018696
+ID000018750
+ID000018765
+ID000018823
+ID000018854
+ID000018870
+ID000018873
+ID000018875
+ID000018877
+ID000018878
+ID000018901
+ID000018902
+ID000018904
+ID000018905
+ID000018936
+ID000018937
+ID000018938
+ID000018939
+ID000018970
+ID000018971
+ID000018972
+ID000019024
+ID000019025
+ID000019038
+ID000019052
+ID000019081
+ID000019082
+ID000019101
+ID000019102
+ID000019145
+ID000019146
+ID000019165
+ID000019166
+ID000019193
+ID000019210
+ID000019264
+ID000019299
+ID000019405
+ID000020309
+ID000020357
+ID000021168
+ID000021170
+ID000021178
+ID000021183
+ID000021261
+ID000021390
+ID000023014
+ID000027519
+ID000028882
+ID000028889
+ID000029205
+ID000029281
+ID000029359
+ID000029365
+ID000029522
+ID000029542
+ID000029554
+ID000029555
+ID000029558
+ID000030669
+ID000030793
+ID000031061
+ID000031069
+ID000031088
+ID000031119
+ID000031141
+ID000031167
+ID000031168
+ID000031207
+ID000031215
+ID000031261
+ID000031278
+ID000031279
+ID000031817
+ID000031820

package/package.json

@@ -2,7 +2,7 @@
   "name": "@datagrok/bio",
   "beta": false,
   "friendlyName": "Bio",
-  "version": "1.4.2",
+  "version": "1.5.3",
   "description": "Bio is a [package](https://datagrok.ai/help/develop/develop#packages) for the [Datagrok](https://datagrok.ai) platform",
   "repository": {
     "type": "git",
@@ -11,11 +11,11 @@
   },
   "dependencies": {
     "@biowasm/aioli": ">=2.4.0",
-    "@datagrok-libraries/bio": "^2.2.0",
-    "@datagrok-libraries/utils": "^0.4.2",
-    "@datagrok-libraries/ml": "^2.0.4",
+    "@datagrok-libraries/bio": "^2.3.1",
+    "@datagrok-libraries/utils": "^1.0.0",
+    "@datagrok-libraries/ml": "^2.0.8",
     "cash-dom": "latest",
-    "datagrok-api": "^1.4.11",
+    "datagrok-api": "^1.4.12",
     "dayjs": "latest",
     "ts-loader": "^9.2.5",
     "typescript": "^4.4.2"
@@ -41,7 +41,8 @@
     "debug-sequences1": "grok publish --rebuild",
     "release-sequences1": "grok publish --rebuild --release",
     "build-sequences1": "webpack",
-    "local-bio": "grok publish local",
+    "debug-local": "grok publish local",
+    "release-local": "grok publish local --release",
     "build": "webpack",
     "debug-sequences1-public": "grok publish public --rebuild",
     "release-sequences1-public": "grok publish public --rebuild --release",

package/setup.cmd

@@ -1,10 +1,19 @@
 cd ../../js-api
 call npm install
 call npm link
+cd ../libraries/utils
+call npm install
+call npm link
+call npm link datagrok-api
+cd ../libraries/ml
+call npm install
+call npm link
+call npm link @datagrok-libraries/utils
 cd ../libraries/bio
 call npm install
 call npm link
+call npm link @datagrok-libraries/utils
 cd ../../packages/Bio
 call npm install
-call npm link datagrok-api @datagrok-libraries/bio
+call npm link datagrok-api @datagrok-libraries/bio @datagrok-libraries/utils @datagrok-libraries/ml
 webpack

package/src/package-test.ts

@@ -7,6 +7,7 @@ import './tests/Palettes-test';
 import './tests/detectors-test';
 import './tests/msa-tests';
 import './tests/sequence-space-test';
+import './tests/activity-cliffs-tests';
 
 export const _package = new DG.Package();
 export {tests};

package/src/package.ts

@@ -2,17 +2,21 @@
 import * as grok from 'datagrok-api/grok';
 import * as ui from 'datagrok-api/ui';
 import * as DG from 'datagrok-api/dg';
-import {SequenceAlignment, Aligned} from './seq_align';
 
 export const _package = new DG.Package();
 
-import {WebLogo} from '@datagrok-libraries/bio/src/viewers/web-logo';
+import {mmSemType} from './const';
+import {WebLogo, SeqColStats} from '@datagrok-libraries/bio/src/viewers/web-logo';
 import {VdRegionsViewer} from './viewers/vd-regions-viewer';
 import {runKalign, testMSAEnoughMemory} from './utils/multiple-sequence-alignment';
+import {SequenceAlignment, Aligned} from './seq_align';
+import {Nucleotides} from '@datagrok-libraries/bio/src/nucleotides';
+import {Aminoacids} from '@datagrok-libraries/bio/src/aminoacids';
 import {convert} from './utils/convert';
-import {TableView} from 'datagrok-api/dg';
-import { getEmbeddingColsNames, sequenceSpace } from './utils/sequence-space';
-import { AvailableMetrics } from '@datagrok-libraries/ml/src/typed-metrics';
+import {getEmbeddingColsNames, sequenceSpace} from './utils/sequence-space';
+import {AvailableMetrics} from '@datagrok-libraries/ml/src/typed-metrics';
+import {getActivityCliffs} from '@datagrok-libraries/ml/src/viewers/activity-cliffs';
+import {sequenceGetSimilarities, drawTooltip} from './utils/sequence-activity-cliffs';
 
 //name: sequenceAlignment
 //input: string alignType {choices: ['Local alignment', 'Global alignment']}
@@ -44,16 +48,35 @@ export function vdRegionViewer() {
   return new VdRegionsViewer();
 }
 
-//top-menu: Bio | Activity Cliffs...
-//name: Activity Cliffs
+//top-menu: Bio | Sequence Activity Cliffs...
+//name: Sequence Activity Cliffs
 //description: detect activity cliffs
 //input: dataframe df [Input data table]
-//input: column smiles {type:categorical; semType: Macromolecule}
+//input: column sequence {semType: Macromolecule}
 //input: column activities
 //input: double similarity = 80 [Similarity cutoff]
 //input: string methodName { choices:["UMAP", "t-SNE", "SPE"] }
-export async function activityCliffs(df: DG.DataFrame, smiles: DG.Column, activities: DG.Column,
+export async function activityCliffs(df: DG.DataFrame, sequence: DG.Column, activities: DG.Column,
   similarity: number, methodName: string): Promise<void> {
+  const axesNames = getEmbeddingColsNames(df);
+  const options = {
+    'SPE': {cycles: 2000, lambda: 1.0, dlambda: 0.0005},
+  };
+  const units = sequence!.tags[DG.TAGS.UNITS];
+  await getActivityCliffs(
+    df,
+    sequence,
+    axesNames,
+    activities,
+    similarity,
+    'Levenshtein',
+    methodName,
+    DG.SEMTYPE.MACROMOLECULE,
+    units,
+    sequenceSpace,
+    sequenceGetSimilarities,
+    drawTooltip,
+    (options as any)[methodName]);
 }
 
 //top-menu: Bio | Sequence Space...
@@ -64,18 +87,24 @@ export async function activityCliffs(df: DG.DataFrame, smiles: DG.Column, activi
 //input: string similarityMetric { choices:["Levenshtein", "Tanimoto"] }
 //input: bool plotEmbeddings = true
 export async function sequenceSpaceTopMenu(table: DG.DataFrame, macroMolecule: DG.Column, methodName: string,
-  similarityMetric: string = 'Levenshtein', plotEmbeddings: boolean) : Promise<void> {
-    const embedColsNames = getEmbeddingColsNames(table);
-    const sequenceSpaceRes = await sequenceSpace(macroMolecule, methodName, similarityMetric, embedColsNames);
-    const embeddings = sequenceSpaceRes.coordinates;
-    for (const col of embeddings)
-      table.columns.add(col);
-    if (plotEmbeddings) {
-      for (let v of grok.shell.views) {
-        if (v.name === table.name)
-          (v as DG.TableView).scatterPlot({x: embedColsNames[0], y: embedColsNames[1]});
-      }
-    } 
+  similarityMetric: string = 'Levenshtein', plotEmbeddings: boolean): Promise<void> {
+  const embedColsNames = getEmbeddingColsNames(table);
+  const chemSpaceParams = {
+    seqCol: macroMolecule,
+    methodName: methodName,
+    similarityMetric: similarityMetric,
+    embedAxesNames: embedColsNames
+  };
+  const sequenceSpaceRes = await sequenceSpace(chemSpaceParams);
+  const embeddings = sequenceSpaceRes.coordinates;
+  for (const col of embeddings)
+    table.columns.add(col);
+  if (plotEmbeddings) {
+    for (const v of grok.shell.views) {
+      if (v.name === table.name)
+        (v as DG.TableView).scatterPlot({x: embedColsNames[0], y: embedColsNames[1]});
+    }
+  }
 };
 
 //top-menu: Bio | MSA...
@@ -100,7 +129,7 @@ export async function compositionAnalysis(): Promise<void> {
   const wl = await col.dataFrame.plot.fromType('WebLogo', {});
 
   for (const v of grok.shell.views) {
-    if (v instanceof TableView && (v as DG.TableView).dataFrame.name === col.dataFrame.name) {
+    if (v instanceof DG.TableView && (v as DG.TableView).dataFrame.name === col.dataFrame.name) {
       (v as DG.TableView).dockManager.dock(wl.root, 'down');
       break;
     }
@@ -122,10 +151,10 @@ function parseMacromolecule(
 //description: Opens FASTA file
 //tags: file-handler
 //meta.ext: fasta, fna, ffn, faa, frn, fa
-//input: string content
+//input: string fileContent
 //output: list tables
 export function importFasta(fileContent: string): DG.DataFrame [] {
-  const regex = /^>(.*)$/gm; // match the line starting with >
+  const regex = /^>(.*)$/gm; // match lines starting with >
   const descriptionsArray = [];
   const sequencesArray: string[] = [];
   let startOfSequence = 0;
@@ -141,6 +170,22 @@ export function importFasta(fileContent: string): DG.DataFrame [] {
   const descriptionsArrayCol = DG.Column.fromStrings('description', descriptionsArray);
   const sequenceCol = DG.Column.fromStrings('sequence', sequencesArray);
   sequenceCol.semType = 'Macromolecule';
+
+  const stats: SeqColStats = WebLogo.getStats(sequenceCol, 5, WebLogo.splitterAsFasta);
+  const seqType = stats.sameLength ? 'SEQ.MSA' : 'SEQ';
+  const alphabetCandidates: [string, Set<string>][] = [
+    ['NT', new Set(Object.keys(Nucleotides.Names))],
+    ['PT', new Set(Object.keys(Aminoacids.Names))],
+  ];
+  // Calculate likelihoods for alphabet_candidates
+  const alphabetCandidatesSim: number[] = alphabetCandidates.map(
+    (c) => WebLogo.getAlphabetSimilarity(stats.freq, c[1]));
+  const maxCos = Math.max(...alphabetCandidatesSim);
+  const alphabet = maxCos > 0.65 ? alphabetCandidates[alphabetCandidatesSim.indexOf(maxCos)][0] : 'UN';
+  sequenceCol.semType = mmSemType;
+  const units: string = `fasta:${seqType}:${alphabet}`;
+  sequenceCol.setTag(DG.TAGS.UNITS, units);
+
   return [DG.DataFrame.fromColumns([
     descriptionsArrayCol,
     sequenceCol,
@@ -153,4 +198,4 @@ export function importFasta(fileContent: string): DG.DataFrame [] {
 //input: column col {semType: Macromolecule}
 export function convertPanel(col: DG.Column): void {
   convert(col);
-}
+}

package/src/tests/activity-cliffs-tests.ts

@@ -0,0 +1,49 @@
+import {after, before, category, expect, expectFloat, test} from '@datagrok-libraries/utils/src/test';
+import * as DG from 'datagrok-api/dg';
+import {createTableView, readDataframe} from './utils';
+import {_package} from '../package-test';
+import {getEmbeddingColsNames, sequenceSpace} from '../utils/sequence-space';
+import {drawTooltip, sequenceGetSimilarities} from '../utils/sequence-activity-cliffs';
+import {getActivityCliffs} from '@datagrok-libraries/ml/src/viewers/activity-cliffs';
+
+
+category('activityCliffs', async () => {
+  let actCliffsTableView: DG.TableView;
+  let actCliffsDf: DG.DataFrame;
+
+  before(async () => {
+    actCliffsTableView = await createTableView('sample_MSA.csv');
+    actCliffsDf = await readDataframe('sample_MSA.csv');
+  });
+
+  test('activityCliffsOpen', async () => {
+    const axesNames = getEmbeddingColsNames(actCliffsDf);
+    const units = actCliffsDf.col('MSA')!.tags[DG.TAGS.UNITS];
+    const options = {
+      'SPE': {cycles: 2000, lambda: 1.0, dlambda: 0.0005},
+    };
+    const scatterPlot = await getActivityCliffs(
+      actCliffsDf,
+      actCliffsDf.col('MSA')!,
+      axesNames,
+      actCliffsDf.col('Activity')!,
+      50,
+      'Levenshtein',
+      't-SNE',
+      DG.SEMTYPE.MACROMOLECULE,
+      units,
+      sequenceSpace,
+      sequenceGetSimilarities,
+      drawTooltip);
+
+    expect(scatterPlot != null, true);
+
+    const cliffsLink = (Array.from(scatterPlot.root.children) as Element[])
+      .filter((it) => it.className === 'ui-btn ui-btn-ok');
+    expect((cliffsLink[0] as HTMLElement).innerText, '101 cliffs');
+  });
+
+  after(async () => {
+    actCliffsTableView.close();
+  });
+});