@datagrok/bio 2.4.19 → 2.4.23

package/package.json

@@ -5,7 +5,7 @@
     "name": "Leonid Stolbov",
     "email": "lstolbov@datagrok.ai"
   },
-  "version": "2.4.19",
+  "version": "2.4.23",
   "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",
@@ -14,10 +14,10 @@
   },
   "dependencies": {
     "@biowasm/aioli": "^3.1.0",
-    "@datagrok-libraries/bio": "^5.29.3",
+    "@datagrok-libraries/bio": "^5.30.0",
     "@datagrok-libraries/chem-meta": "^1.0.1",
     "@datagrok-libraries/ml": "^6.3.23",
-    "@datagrok-libraries/tutorials": "^1.3.1",
+    "@datagrok-libraries/tutorials": "^1.3.2",
     "@datagrok-libraries/utils": "^2.1.3",
     "cash-dom": "^8.0.0",
     "css-loader": "^6.7.3",

package/src/demo/bio01-similarity-diversity.ts

@@ -9,7 +9,7 @@ import {handleError} from './utils';
 import {SequenceDiversityViewer} from '../analysis/sequence-diversity-viewer';
 import {SequenceSimilarityViewer} from '../analysis/sequence-similarity-viewer';
 
-const dataFn: string = 'data/sample_FASTA_DNA.csv';
+const dataFn: string = 'data/sample_FASTA_PT_activity.csv';
 
 export async function demoBio01UI() {
   let view: DG.TableView;
@@ -27,8 +27,12 @@ export async function demoBio01UI() {
         df = await _package.files.readCsv(dataFn);
         view = grok.shell.addTableView(df);
 
-        view.grid.columns.byName('id')!.width = 0;
-        view.grid.columns.byName('sequence')!.width = 500;
+        view.grid.columns.byName('cluster')!.visible = false;
+        view.grid.columns.byName('sequence_id')!.visible = false;
+        view.grid.columns.byName('sequence')!.width = 300;
+        view.grid.columns.byName('activity')!.visible = false;
+        view.grid.columns.byName('is_cliff')!.visible = false;
+
         // TODO: Fix column width
       }, {
         description: `Load dataset with macromolecules of 'fasta' notation, 'DNA' alphabet.`,

package/src/demo/bio01a-hierarchical-clustering-and-sequence-space.ts

@@ -11,7 +11,7 @@ import {getDendrogramService, IDendrogramService} from '@datagrok-libraries/bio/
 import {demoSequenceSpace, handleError} from './utils';
 import {DemoScript} from '@datagrok-libraries/tutorials/src/demo-script';
 
-const dataFn = 'data/sample_FASTA_DNA.csv';
+const dataFn = 'data/sample_FASTA_PT_activity.csv';
 const seqColName = 'sequence';
 
 export async function demoBio01aUI() {
@@ -21,7 +21,7 @@ export async function demoBio01aUI() {
   let df: DG.DataFrame;
   let spViewer: DG.ScatterPlotViewer;
 
-  const method: string = 'UMAP';
+  const dimRedMethod: string = 'UMAP';
   const idRows: { [id: number]: number } = {};
   const embedCols: { [colName: string]: DG.Column<number> } = {};
 
@@ -38,6 +38,9 @@ export async function demoBio01aUI() {
         ]);
         view = grok.shell.addTableView(df);
         view.grid.props.rowHeight = 22;
+        view.grid.columns.byName('cluster')!.visible = false;
+        view.grid.columns.byName('sequence')!.width = 200;
+        view.grid.columns.byName('is_cliff')!.visible = false;
 
         grok.shell.windows.showContextPanel = false;
         grok.shell.windows.showProperties = false;
@@ -46,7 +49,7 @@ export async function demoBio01aUI() {
         delay: 2000,
       })
       .step('Build sequence space', async () => {
-        spViewer = await demoSequenceSpace(view, df, seqColName, method);
+        spViewer = await demoSequenceSpace(view, df, seqColName, dimRedMethod);
       }, {
         description: `Reduce sequence space dimensionality to display on 2D representation.`,
         delay: 2000
@@ -71,7 +74,10 @@ export async function demoBio01aUI() {
         delay: 2000,
       })
       .step('Select a bunch of sequences', async () => {
-        df.selection.init((idx: number) => [21, 9, 58].includes(idx));
+        const seqIdCol: DG.Column<string> = df.getCol('sequence_id');
+        df.selection.init((rowI: number) => {
+          return ['c0_seq120', 'c0_seq105', 'c0_seq121', 'c0_seq93'].includes(seqIdCol.get(rowI)!);
+        });
         df.currentRowIdx = 27;
       }, {
         description: 'Selecting a group of rows from a data frame to show their similarity and proximity to each other on a viewer..',

package/src/demo/bio01b-hierarchical-clustering-and-activity-cliffs.ts

@@ -13,7 +13,7 @@ import {getDendrogramService, IDendrogramService} from '@datagrok-libraries/bio/
 import {handleError} from './utils';
 import {DemoScript} from '@datagrok-libraries/tutorials/src/demo-script';
 
-const dataFn: string = 'samples/sample_FASTA.csv';
+const dataFn: string = 'data/sample_FASTA_PT_activity.csv';
 
 export async function demoBio01bUI() {
   let treeHelper: ITreeHelper;
@@ -23,7 +23,7 @@ export async function demoBio01bUI() {
   let view: DG.TableView;
   let activityCliffsViewer: DG.ScatterPlotViewer;
 
-  const method: string = 'UMAP';
+  const dimRedMethod: string = 'UMAP';
   const idRows: { [id: number]: number } = {};
 
   try {
@@ -43,10 +43,9 @@ export async function demoBio01bUI() {
 
         view = grok.shell.addTableView(df);
         view.grid.props.rowHeight = 22;
-        const uniProtKbGCol = view.grid.columns.byName('UniProtKB')!;
-        uniProtKbGCol.width = 75;
-        const lengthGCol = view.grid.columns.byName('Length')!;
-        lengthGCol.width = 0;
+        view.grid.columns.byName('cluster')!.visible = false;
+        view.grid.columns.byName('sequence')!.width = 300;
+        view.grid.columns.byName('is_cliff')!.visible = false;
       }, {
         description: 'Load dataset with macromolecules of \'fasta\' notation, \'DNA\' alphabet.',
         delay: 2000,
@@ -54,7 +53,7 @@ export async function demoBio01bUI() {
       .step('Find activity cliffs', async () => {
         activityCliffsViewer = (await activityCliffs(
           df, df.getCol('Sequence'), df.getCol('Activity'),
-          80, method)) as DG.ScatterPlotViewer;
+          80, dimRedMethod)) as DG.ScatterPlotViewer;
         view.dockManager.dock(activityCliffsViewer, DG.DOCK_TYPE.RIGHT, null, 'Activity Cliffs', 0.35);
 
         // Show grid viewer with the cliffs
@@ -86,7 +85,7 @@ export async function demoBio01bUI() {
         //cliffsDfGrid.dataFrame.currentRowIdx = -1; // reset
         const cliffsDfGrid: DG.Grid = activityCliffsViewer.dataFrame.temp[acTEMPS.cliffsDfGrid];
         //cliffsDfGrid.dataFrame.selection.init((i) => i == currentCliffIdx);
-        cliffsDfGrid.dataFrame.currentRowIdx = 0;
+        if (cliffsDfGrid.dataFrame.rowCount > 0) cliffsDfGrid.dataFrame.currentRowIdx = 0;
         //cliffsDfGrid.dataFrame.selection.set(currentCliffIdx, true, true);
 
         // /* workaround to select rows of the cliff */

package/src/demo/bio05-helm-msa-sequence-space.ts

@@ -22,6 +22,7 @@ export async function demoBio05UI(): Promise<void> {
 
   const helmColName: string = 'HELM';
   const msaHelmColName: string = 'msa(HELM)';
+  const dimRedMethod: string = 'UMAP';
 
   try {
     const demoScript = new DemoScript(
@@ -37,7 +38,7 @@ export async function demoBio05UI(): Promise<void> {
         description: 'Load dataset with macromolecules of \'Helm\' notation.',
         delay: 2000,
       })
-      .step('Align paptides with non-natural aminoacids with PepSeA', async () => {
+      .step('Align peptides with non-natural aminoacids with PepSeA', async () => {
         helmCol = df.getCol(helmColName);
         const method: string = pepseaMethods[0];
         const gapOpen: number = 1.53;
@@ -50,9 +51,8 @@ export async function demoBio05UI(): Promise<void> {
         delay: 2000,
       })
       .step('Build sequence space', async () => {
-        const method: string = 'UMAP';
         ssViewer = (await sequenceSpaceTopMenu(df, msaHelmCol,
-          'UMAP', StringMetricsNames.Levenshtein, true)) as DG.ScatterPlotViewer;
+          dimRedMethod, StringMetricsNames.Levenshtein, true)) as DG.ScatterPlotViewer;
         view.dockManager.dock(ssViewer, DG.DOCK_TYPE.RIGHT, null, 'Sequence Space', 0.35);
       }, {
         description: 'Reduce sequence space dimensionality to display on 2D representation.',

package/src/demo/utils.ts

@@ -52,18 +52,6 @@ export async function demoSequenceSpace(
         embedCol.init((rowI) => { return embedColData[rowI]; });
       }
 
-      const rowCount: number = df.rowCount;
-      const idCol: DG.Column = df.getCol('id');
-      for (let idRowI = 0; idRowI < rowCount; idRowI++) {
-        const id = idCol.get(idRowI);
-        //idRows[id] = idRowI;
-      }
-
-      for (const embedColName of Object.values(EMBED_COL_NAMES)) {
-        const embedCol: DG.Column<number> = df.getCol(embedColName);
-        //embedCols[embedColName] = embedCol;
-      }
-
       const t3: number = Date.now();
       _package.logger.debug('MLB: MlbVrSpaceBrowser.buildView(), postprocess reduceDimensionality ' +
         `ET: ${((t3 - t2) / 1000)} s`);

package/src/package-test.ts

@@ -22,6 +22,7 @@ import './tests/substructure-filters-tests';
 import './tests/pepsea-tests';
 import './tests/viewers';
 import './tests/units-handler-tests';
+import './tests/to-atomic-level-tests';
 import './tests/mm-distance-tests';
 
 // Tests hanging github CI

package/src/package.ts

@@ -9,7 +9,7 @@ import {
 } from './utils/cell-renderer';
 import {VdRegionsViewer} from './viewers/vd-regions-viewer';
 import {SequenceAlignment} from './seq_align';
-import {getEmbeddingColsNames, sequenceSpaceByFingerprints} from './analysis/sequence-space';
+import {getEmbeddingColsNames, sequenceSpaceByFingerprints, getSequenceSpace} from './analysis/sequence-space';
 import {getActivityCliffs} from '@datagrok-libraries/ml/src/viewers/activity-cliffs';
 import {
   createLinesGrid,
@@ -290,19 +290,23 @@ export async function activityCliffs(df: DG.DataFrame, macroMolecule: DG.Column,
     'separator': macroMolecule.getTag(bioTAGS.separator),
     'alphabet': macroMolecule.getTag(bioTAGS.alphabet),
   };
+  const uh = new UnitsHandler(macroMolecule);
+  let columnDistanceMetric = 'Tanimoto';
+  if (uh.isFasta())
+    columnDistanceMetric = uh.getDistanceFunctionName();
   const sp = await getActivityCliffs(
     df,
     macroMolecule,
     null,
     axesNames,
-    'Activity cliffs',
+    'Activity cliffs', //scatterTitle
     activities,
     similarity,
-    'Tanimoto',
+    columnDistanceMetric, //similarityMetric
     methodName,
     DG.SEMTYPE.MACROMOLECULE,
     tags,
-    sequenceSpaceByFingerprints,
+    getSequenceSpace,
     getChemSimilaritiesMatrix,
     createTooltipElement,
     createPropPanelElement,
@@ -353,7 +357,7 @@ export async function sequenceSpaceTopMenu(table: DG.DataFrame, macroMolecule: D
     embedAxesNames: embedColsNames,
     options: options
   };
-  const sequenceSpaceRes = await sequenceSpaceByFingerprints(chemSpaceParams);
+  const sequenceSpaceRes = await getSequenceSpace(chemSpaceParams);
   const embeddings = sequenceSpaceRes.coordinates;
   for (const col of embeddings) {
     const listValues = col.toList();
@@ -407,9 +411,15 @@ export async function toAtomicLevel(df: DG.DataFrame, macroMolecule: DG.Column):
   }
   if (!checkInputColumnUI(macroMolecule, 'To Atomic Level'))
     return;
-  const monomersLibFile = await _package.files.readAsText(HELM_CORE_LIB_FILENAME);
-  const monomersLibObject: any[] = JSON.parse(monomersLibFile);
-  await _toAtomicLevel(df, macroMolecule, monomersLibObject);
+  const monomerLib: IMonomerLib = (await getMonomerLibHelper()).getBioLib();
+  const atomicLevelRes = await _toAtomicLevel(df, macroMolecule, monomerLib);
+  if (atomicLevelRes.col !== null) {
+    df.columns.add(atomicLevelRes.col, true);
+    await grok.data.detectSemanticTypes(df);
+  }
+
+  if (atomicLevelRes.warnings && atomicLevelRes.warnings.length > 0)
+    grok.shell.warning(ui.list(atomicLevelRes.warnings));
 }
 
 //top-menu: Bio | Alignment | MSA...

package/src/tests/converters-test.ts

@@ -53,9 +53,9 @@ Y-N-R-Q-W-Y-V
 M-K-P-S-E-Y-V
 `,
     helmPt: `seq
-PEPTIDE1{F.W.P.H.E.Y}$$$
-PEPTIDE1{Y.N.R.Q.W.Y.V}$$$
-PEPTIDE1{M.K.P.S.E.Y.V}$$$
+PEPTIDE1{F.W.P.H.E.Y}$$$$
+PEPTIDE1{Y.N.R.Q.W.Y.V}$$$$
+PEPTIDE1{M.K.P.S.E.Y.V}$$$$
 `,
     fastaDna: `seq
 ACGTC
@@ -68,9 +68,9 @@ C/A/G/T/G/T
 T/T/C/A/A/C
 `,
     helmDna: `seq
-DNA1{D(A)P.D(C)P.D(G)P.D(T)P.D(C)P}$$$
-DNA1{D(C)P.D(A)P.D(G)P.D(T)P.D(G)P.D(T)P}$$$
-DNA1{D(T)P.D(T)P.D(C)P.D(A)P.D(A)P.D(C)P}$$$
+DNA1{D(A)P.D(C)P.D(G)P.D(T)P.D(C)P}$$$$
+DNA1{D(C)P.D(A)P.D(G)P.D(T)P.D(G)P.D(T)P}$$$$
+DNA1{D(T)P.D(T)P.D(C)P.D(A)P.D(A)P.D(C)P}$$$$
 `,
     fastaRna: `seq
 ACGUC
@@ -83,9 +83,9 @@ C*A*G*U*G*U
 U*U*C*A*A*C
 `,
     helmRna: `seq
-RNA1{R(A)P.R(C)P.R(G)P.R(U)P.R(C)P}$$$
-RNA1{R(C)P.R(A)P.R(G)P.R(U)P.R(G)P.R(U)P}$$$
-RNA1{R(U)P.R(U)P.R(C)P.R(A)P.R(A)P.R(C)P}$$$
+RNA1{R(A)P.R(C)P.R(G)P.R(U)P.R(C)P}$$$$
+RNA1{R(C)P.R(A)P.R(G)P.R(U)P.R(G)P.R(U)P}$$$$
+RNA1{R(U)P.R(U)P.R(C)P.R(A)P.R(A)P.R(C)P}$$$$
 `,
     fastaGaps: `seq
 FW-PH-EYY
@@ -98,9 +98,9 @@ F/Y/N/R/Q/W/Y/V/
 F/K/P//Q//S/E/Y/V
 `,
     helmGaps: `seq
-PEPTIDE1{F.W.*.P.H.*.E.Y.Y}$$$
-PEPTIDE1{F.Y.N.R.Q.W.Y.V.*}$$$
-PEPTIDE1{F.K.P.*.Q.*.S.E.Y.V}$$$
+PEPTIDE1{F.W.*.P.H.*.E.Y.Y}$$$$
+PEPTIDE1{F.Y.N.R.Q.W.Y.V.*}$$$$
+PEPTIDE1{F.K.P.*.Q.*.S.E.Y.V}$$$$
 `,
 
     fastaUn: `seq
@@ -114,24 +114,24 @@ meI-hHis-Aca-Cys_SEt-T-dK-Thr_PO3H2-Aca-Tyr_PO3H2
 Lys_Boc-hHis-Aca-Cys_SEt-T-dK-Thr_PO3H2-Aca-Tyr_PO3H2
 `,
     helmUn: `seq
-PEPTIDE1{meI.hHis.Aca.N.T.dE.Thr_PO3H2.Aca.D}$$$
-PEPTIDE1{meI.hHis.Aca.Cys_SEt.T.dK.Thr_PO3H2.Aca.Tyr_PO3H2}$$$
-PEPTIDE1{Lys_Boc.hHis.Aca.Cys_SEt.T.dK.Thr_PO3H2.Aca.Tyr_PO3H2}$$$
+PEPTIDE1{meI.hHis.Aca.N.T.dE.Thr_PO3H2.Aca.D}$$$$
+PEPTIDE1{meI.hHis.Aca.Cys_SEt.T.dK.Thr_PO3H2.Aca.Tyr_PO3H2}$$$$
+PEPTIDE1{Lys_Boc.hHis.Aca.Cys_SEt.T.dK.Thr_PO3H2.Aca.Tyr_PO3H2}$$$$
 `,
     helmLoneDeoxyribose: `seq
-DNA1{D(A).D(C).D(G).D(T).D(C)}$$$
-DNA1{D(C).D(A).D(G).D(T).D(G).D(T)P}$$$
-DNA1{D(T).D(T).D(C).D(A).D(A).D(C)P}$$$
+DNA1{D(A).D(C).D(G).D(T).D(C)}$$$$
+DNA1{D(C).D(A).D(G).D(T).D(G).D(T)P}$$$$
+DNA1{D(T).D(T).D(C).D(A).D(A).D(C)P}$$$$
 `,
     helmLoneRibose: `seq
-RNA1{R(A).R(C).R(G).R(U).R(C)}$$$
-RNA1{R(C).R(A).R(G).R(U).R(G).R(U)P}$$$
-RNA1{R(U).R(U).R(C).R(A).R(A).R(C)P}$$$
+RNA1{R(A).R(C).R(G).R(U).R(C)}$$$$
+RNA1{R(C).R(A).R(G).R(U).R(G).R(U)P}$$$$
+RNA1{R(U).R(U).R(C).R(A).R(A).R(C)P}$$$$
 `,
     helmLonePhosphorus: `seq
-RNA1{P.P.R(A)P.R(C)P.R(G)P.R(U)P.R(C)P}$$$
-RNA1{P.P.R(C)P.R(A)P.P.R(G)P.R(U)P.R(G)P.R(U)P}$$$
-RNA1{P.R(U)P.R(U)P.R(C)P.R(A)P.R(A)P.R(C)P.P.P}$$$
+RNA1{P.P.R(A)P.R(C)P.R(G)P.R(U)P.R(C)P}$$$$
+RNA1{P.P.R(C)P.R(A)P.P.R(G)P.R(U)P.R(G)P.R(U)P}$$$$
+RNA1{P.R(U)P.R(U)P.R(C)P.R(A)P.R(A)P.R(C)P.P.P}$$$$
 `,
   };
 

package/src/tests/to-atomic-level-tests.ts

@@ -0,0 +1,187 @@
+/* Do not change these import lines to match external modules in webpack configuration */
+import * as grok from 'datagrok-api/grok';
+import * as ui from 'datagrok-api/ui';
+import * as DG from 'datagrok-api/dg';
+
+import {before, after, category, test, expectArray} from '@datagrok-libraries/utils/src/test';
+
+import {getMonomerLibHelper, toAtomicLevel} from '../package';
+import {_toAtomicLevel} from '@datagrok-libraries/bio/src/monomer-works/to-atomic-level';
+import {IMonomerLib} from '@datagrok-libraries/bio/src/types/index';
+import {IMonomerLibHelper} from '@datagrok-libraries/bio/src/monomer-works/monomer-utils';
+import {LIB_STORAGE_NAME} from '../utils/monomer-lib';
+
+const appPath = 'System:AppData/Bio';
+const fileSource = new DG.FileSource(appPath);
+
+const testNames: { [k: string]: string } = {
+  PT: 'peptides fasta',
+  DNA: 'dna fasta',
+  MSA: 'msa separator',
+};
+
+const inputPath: { [k: string]: string } = {
+  PT: 'tests/to-atomic-level-peptides-fasta-input.csv',
+  DNA: 'tests/to-atomic-level-dna-fasta-input.csv',
+  MSA: 'tests/to-atomic-level-msa-separator-input.csv',
+};
+
+const outputPath: { [k: string]: string } = {
+  PT: 'tests/to-atomic-level-peptides-output.csv',
+  DNA: 'tests/to-atomic-level-dna-output.csv',
+  MSA: 'tests/to-atomic-level-msa-output.csv',
+};
+
+const inputColName = 'sequence';
+const outputColName = 'molfile(sequence)';
+
+category('toAtomicLevel', async () => {
+  const sourceDf: { [key: string]: DG.DataFrame } = {};
+  const targetDf: { [key: string]: DG.DataFrame } = {};
+
+  let monomerLibHelper: IMonomerLibHelper;
+  /** Backup actual user's monomer libraries settings */
+  let userLibrariesSettings: any = null;
+
+  before(async () => {
+    monomerLibHelper = await getMonomerLibHelper();
+    userLibrariesSettings = await grok.dapi.userDataStorage.get(LIB_STORAGE_NAME, true);
+    // Clear settings to test default
+    await grok.dapi.userDataStorage.put(LIB_STORAGE_NAME, {}, true);
+    await monomerLibHelper.loadLibraries(true);
+
+    for (const key in testNames) {
+      sourceDf[key] = await fileSource.readCsv(inputPath[key]);
+      await grok.data.detectSemanticTypes(sourceDf[key]);
+      targetDf[key] = await fileSource.readCsv(outputPath[key]);
+    }
+  });
+
+  after(async () => {
+    await grok.dapi.userDataStorage.put(LIB_STORAGE_NAME, userLibrariesSettings, true);
+    await monomerLibHelper.loadLibraries(true);
+  });
+
+  async function getTestResult(source: DG.DataFrame, target: DG.DataFrame): Promise<void> {
+    const inputCol = source.getCol(inputColName);
+    await toAtomicLevel(source, inputCol);
+    const obtainedCol = source.getCol(outputColName);
+    const expectedCol = target.getCol(outputColName);
+    const obtainedArray = [...obtainedCol.values()];
+    const expectedArray = [...expectedCol.values()];
+    expectArray(obtainedArray, expectedArray);
+  }
+
+  for (const key in testNames) {
+    test(`${testNames[key]}`, async () => {
+      await getTestResult(sourceDf[key], targetDf[key]);
+    }, {skipReason: 'GROK-13100'});
+  }
+
+  enum csvTests {
+    fastaDna = 'fastaDna',
+    fastaRna = 'fastaRna',
+    fastaPt = 'fastaPt',
+
+    separatorDna = 'separatorDna',
+    separatorRna = 'separatorRna',
+    separatorPt = 'separatorPt',
+    separatorUn = 'separatorUn',
+
+    helm = 'helm',
+  }
+
+  const csvData: { [key in csvTests]: string } = {
+    [csvTests.fastaDna]: `seq
+ACGTC
+CAGTGT
+TTCAAC
+`,
+    [csvTests.fastaRna]: `seq
+ACGUC
+CAGUGU
+UUCAAC
+`,
+    [csvTests.fastaPt]: `seq
+FWPHEY
+YNRQWYV
+MKPSEYV
+`,
+    [csvTests.separatorDna]: `seq
+A/C/G/T/C
+C/A/G/T/G/T
+T/T/C/A/A/C
+`,
+    [csvTests.separatorRna]: `seq
+A*C*G*U*C
+C*A*G*U*G*U
+U*U*C*A*A*C
+`,
+    [csvTests.separatorPt]: `seq
+F-W-P-H-E-Y
+Y-N-R-Q-W-Y-V
+M-K-P-S-E-Y-V
+`,
+    [csvTests.separatorUn]: `seq
+meI-hHis-Aca-N-T-dE-Thr_PO3H2-Aca-D
+meI-hHis-Aca-Cys_SEt-T-dK-Thr_PO3H2-Aca-Tyr_PO3H2
+Lys_Boc-hHis-Aca-Cys_SEt-T-dK-Thr_PO3H2-Aca-Tyr_PO3H2
+`,
+
+    [csvTests.helm]: `seq
+PEPTIDE1{meI.D-gGlu.Aca.N.T.dE.Thr_PO3H2.Aca.D}$$$
+PEPTIDE1{meI.hHis.Aca.Cys_SEt.T.dK.Thr_PO3H2.Aca.Tyr_PO3H2}$$$
+PEPTIDE1{Lys_Boc.hHis.Aca.Cys_SEt.T.dK.Thr_PO3H2.Aca.Tyr_PO3H2}$$$
+`,
+  };
+
+  /** Also detects semantic types
+   * @param {string} key
+   * @return {Promise<DG.DataFrame>}
+   */
+  async function readCsv(key: csvTests): Promise<DG.DataFrame> {
+    // Always recreate test data frame from CSV for reproducible detector behavior in tests.
+    const csv: string = csvData[key];
+    const df: DG.DataFrame = DG.DataFrame.fromCsv(csv);
+    await grok.data.detectSemanticTypes(df);
+    return df;
+  }
+
+  test('fastaDna', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.fastaDna), 'seq', monomerLibHelper);
+  });
+
+  test('fastaRna', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.fastaRna), 'seq', monomerLibHelper);
+  });
+
+  test('fastaPt', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.fastaPt), 'seq', monomerLibHelper);
+  });
+
+  test('separatorDna', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.separatorDna), 'seq', monomerLibHelper);
+  });
+
+  test('separatorDna', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.separatorRna), 'seq', monomerLibHelper);
+  });
+
+  test('separatorPt', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.separatorPt), 'seq', monomerLibHelper);
+  });
+
+  test('separatorUn', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.separatorUn), 'seq', monomerLibHelper);
+  });
+
+  test('helm', async () => {
+    await _testToAtomicLevel(await readCsv(csvTests.helm), 'seq', monomerLibHelper);
+  });
+});
+
+async function _testToAtomicLevel(df: DG.DataFrame, seqColName: string = 'seq', monomerLibHelper: IMonomerLibHelper) {
+  const seqCol: DG.Column<string> = df.getCol(seqColName);
+  const monomerLib: IMonomerLib = monomerLibHelper.getBioLib();
+  const resCol = await _toAtomicLevel(df, seqCol, monomerLib);
+}