@datagrok/peptides 1.25.1 → 1.26.1

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@datagrok/peptides",
   "friendlyName": "Peptides",
-  "version": "1.25.1",
+  "version": "1.26.1",
   "author": {
     "name": "Davit Rizhinashvili",
     "email": "drizhinashvili@datagrok.ai"

package/src/model.ts

@@ -511,53 +511,54 @@ export class PeptidesModel {
       descritionsHost,
     ], 'css-gap-small'));
 
-    if (filterAndSelectionBs.anyTrue) {
-      acc.addPane('Actions', () => {
-        try {
-          const newView = ui.label('New view');
-          $(newView).addClass('d4-link-action');
-          newView.onclick = (): string => trueModel.createNewView();
-          newView.onmouseover = (ev): void =>
-            ui.tooltip.show('Creates a new view from current selection', ev.clientX + 5, ev.clientY + 5);
-          if (trueLSTViewer === null)
-            return ui.divV([newView]);
-
-
-          const newCluster = ui.label('New cluster');
-          $(newCluster).addClass('d4-link-action');
-          newCluster.onclick = (): void => {
-            if (trueLSTViewer === null)
-              throw new Error('Logo summary table viewer is not found');
-
-
-            trueLSTViewer.clusterFromSelection();
-          };
-          newCluster.onmouseover = (ev): void =>
-            ui.tooltip.show('Creates a new cluster from selection', ev.clientX + 5, ev.clientY + 5);
-
-          const removeCluster = ui.label('Remove cluster');
-          $(removeCluster).addClass('d4-link-action');
-          removeCluster.onclick = (): void => {
-            const lstViewer = trueModel.findViewer(VIEWER_TYPE.LOGO_SUMMARY_TABLE) as LogoSummaryTable | null;
-            if (lstViewer === null)
-              throw new Error('Logo summary table viewer is not found');
-
-
-            lstViewer.removeCluster();
-          };
-          removeCluster.onmouseover = (ev): void =>
-            ui.tooltip.show('Removes currently selected custom cluster', ev.clientX + 5, ev.clientY + 5);
-          removeCluster.style.visibility = trueLSTViewer.clusterSelection[CLUSTER_TYPE.CUSTOM].length === 0 ? 'hidden' :
-            'visible';
-
-          return ui.divV([newView, newCluster, removeCluster]);
-        } catch (e) {
-          const errorDiv = ui.divText('Error in Actions');
-          ui.tooltip.bind(errorDiv, String(e));
-          return errorDiv;
-        }
-      }, true);
-    }
+    // ACTIONS PANE temporarily disabled
+    // if (filterAndSelectionBs.anyTrue) {
+    //   acc.addPane('Actions', () => {
+    //     try {
+    //       const newView = ui.label('New view');
+    //       $(newView).addClass('d4-link-action');
+    //       newView.onclick = (): string => trueModel.createNewView();
+    //       newView.onmouseover = (ev): void =>
+    //         ui.tooltip.show('Creates a new view from current selection', ev.clientX + 5, ev.clientY + 5);
+    //       if (trueLSTViewer === null)
+    //         return ui.divV([newView]);
+
+
+    //       const newCluster = ui.label('New cluster');
+    //       $(newCluster).addClass('d4-link-action');
+    //       newCluster.onclick = (): void => {
+    //         if (trueLSTViewer === null)
+    //           throw new Error('Logo summary table viewer is not found');
+
+
+    //         trueLSTViewer.clusterFromSelection();
+    //       };
+    //       newCluster.onmouseover = (ev): void =>
+    //         ui.tooltip.show('Creates a new cluster from selection', ev.clientX + 5, ev.clientY + 5);
+
+    //       const removeCluster = ui.label('Remove cluster');
+    //       $(removeCluster).addClass('d4-link-action');
+    //       removeCluster.onclick = (): void => {
+    //         const lstViewer = trueModel.findViewer(VIEWER_TYPE.LOGO_SUMMARY_TABLE) as LogoSummaryTable | null;
+    //         if (lstViewer === null)
+    //           throw new Error('Logo summary table viewer is not found');
+
+
+    //         lstViewer.removeCluster();
+    //       };
+    //       removeCluster.onmouseover = (ev): void =>
+    //         ui.tooltip.show('Removes currently selected custom cluster', ev.clientX + 5, ev.clientY + 5);
+    //       removeCluster.style.visibility = trueLSTViewer.clusterSelection[CLUSTER_TYPE.CUSTOM].length === 0 ? 'hidden' :
+    //         'visible';
+
+    //       return ui.divV([newView, newCluster, removeCluster]);
+    //     } catch (e) {
+    //       const errorDiv = ui.divText('Error in Actions');
+    //       ui.tooltip.bind(errorDiv, String(e));
+    //       return errorDiv;
+    //     }
+    //   }, true);
+    // }
 
     // Get the source of the bitset change and find viewers that share the same parameters as source
     let requestSource: SARViewer | LogoSummaryTable | PeptidesSettings | null = trueModel.settings;
@@ -587,27 +588,27 @@ export class PeptidesModel {
       v !== null && areParametersEqual(requestSource!, v) && (v !== trueModel.settings || trueModel.isInitialized);
     const panelDataSources = viewers.filter(notEmpty);
     panelDataSources.push(requestSource);
-    const combinedBitset: DG.BitSet | null = DG.BitSet.create(trueModel.df.rowCount);
-    for (const panelDataSource of panelDataSources) {
-      const bitset =
-        (panelDataSource === this.settings) ? getSelectionBitset(this.webLogoSelection, this.monomerPositionStats!) :
-          (panelDataSource instanceof LogoSummaryTable) ?
-            getSelectionBitset(panelDataSource.clusterSelection, panelDataSource.clusterStats) :
-            (panelDataSource instanceof SARViewer) ?
-              getSelectionBitset(panelDataSource.mutationCliffsSelection,
-                mutationCliffsToMaskInfo(panelDataSource.mutationCliffs ?? new Map(), trueModel.df.rowCount)) :
-              null;
-      if (bitset !== null)
-        combinedBitset.or(bitset);
-
-
-      if (panelDataSource instanceof MonomerPosition) {
-        const invariantMapSelectionBitset = getSelectionBitset(panelDataSource.invariantMapSelection,
-          panelDataSource.monomerPositionStats);
-        if (invariantMapSelectionBitset !== null)
-          combinedBitset.or(invariantMapSelectionBitset);
-      }
-    }
+    //const combinedBitset: DG.BitSet | null = DG.BitSet.create(trueModel.df.rowCount);
+    // for (const panelDataSource of panelDataSources) {
+    //   const bitset =
+    //     (panelDataSource === this.settings) ? getSelectionBitset(this.webLogoSelection, this.monomerPositionStats!) :
+    //       (panelDataSource instanceof LogoSummaryTable) ?
+    //         getSelectionBitset(panelDataSource.clusterSelection, panelDataSource.clusterStats) :
+    //         (panelDataSource instanceof SARViewer) ?
+    //           getSelectionBitset(panelDataSource.mutationCliffsSelection,
+    //             mutationCliffsToMaskInfo(panelDataSource.mutationCliffs ?? new Map(), trueModel.df.rowCount)) :
+    //           null;
+    //   if (bitset !== null)
+    //     combinedBitset.or(bitset);
+
+
+    //   if (panelDataSource instanceof MonomerPosition) {
+    //     const invariantMapSelectionBitset = getSelectionBitset(panelDataSource.invariantMapSelection,
+    //       panelDataSource.monomerPositionStats);
+    //     if (invariantMapSelectionBitset !== null)
+    //       combinedBitset.or(invariantMapSelectionBitset);
+    //   }
+    // }
 
     const sarViewer = requestSource as any as SARViewer | LogoSummaryTable;
     if (requestSource !== trueModel.settings && !(sarViewer instanceof LogoSummaryTable) &&
@@ -630,7 +631,6 @@ export class PeptidesModel {
     acc.addPane('Distribution', () => {
       try {
         return getDistributionWidget(trueModel.df, {
-          peptideSelection: combinedBitset,
           columns: isModelSource ? trueModel.settings!.columns ?? {} :
             (requestSource as PeptideViewer).getAggregationColumns(),
           activityCol: isModelSource ? trueModel.getScaledActivityColumn()! :
@@ -869,7 +869,8 @@ export class PeptidesModel {
       }
     }
 
-    return DG.BitSet.fromBytes(combinedSelection.buffer.buffer as ArrayBuffer, combinedSelection.length);
+    const res = DG.BitSet.fromBytes(combinedSelection.buffer.buffer as ArrayBuffer, combinedSelection.length);
+    return this.df.filter.anyFalse ? res.and(this.df.filter) : res;
   }
 
 
@@ -1402,6 +1403,13 @@ export class PeptidesModel {
       await DG.delay(500);
       this._mclViewer = this.findViewer(VIEWER_TYPE.MCL) as MCLViewer;
       await this._mclViewer.initPromise;
+      // after waiting for init promise, it can be resolved by just removing the viewer, so check again
+      this._mclViewer = this.findViewer(VIEWER_TYPE.MCL) as MCLViewer | null;
+      if (!this._mclViewer || !this._mclViewer.isDetached) {
+        mclAdditionSub?.unsubscribe();
+        columnAddedSub?.unsubscribe();
+        return;
+      }
       const lstViewer = this.findViewer(VIEWER_TYPE.LOGO_SUMMARY_TABLE) as LogoSummaryTable | null;
       if (lstViewer) { // beware, this is accessing private things
         lstViewer._clusterStats = null;

package/src/package-api.ts

@@ -48,6 +48,13 @@ export namespace funcs {
     return await grok.functions.call('Peptides:MostPotentResidues', {});
   }
 
+  /**
+  Mutation Cliffs Line Chart
+  */
+  export async function mutationCliffs(): Promise<any> {
+    return await grok.functions.call('Peptides:MutationCliffs', {});
+  }
+
   export async function logoSummaryTable(): Promise<any> {
     return await grok.functions.call('Peptides:LogoSummaryTable', {});
   }

package/src/package.g.ts

@@ -48,6 +48,15 @@ export function mostPotentResidues() : any {
   return PackageFunctions.mostPotentResidues();
 }
 
+//name: Sequence Mutation Cliffs
+//description: Mutation Cliffs Line Chart
+//tags: viewer
+//output: viewer result
+//meta.icon: files/icons/sequence-statistics-viewer.svg
+export function mutationCliffs() : any {
+  return PackageFunctions.mutationCliffs();
+}
+
 //name: Logo Summary Table
 //tags: viewer
 //output: viewer result

package/src/package.ts

@@ -18,6 +18,7 @@ import {ClusterMaxActivityViewer} from './viewers/cluster-max-activity-viewer';
 import {LSTPieChartRenderer} from './utils/cell-renderer';
 import {PeptideUtils} from './peptideUtils';
 import {SequencePositionStatsViewer} from './viewers/position-statistics-viewer';
+import {MutationCliffsViewer} from './viewers/mutation-cliffs-viewer';
 
 let monomerWorks: MonomerWorks | null = null;
 let treeHelper: ITreeHelper;
@@ -193,6 +194,17 @@ export class PackageFunctions {
     return new MostPotentResidues();
   }
 
+  @grok.decorators.func({
+    meta: {icon: 'files/icons/sequence-statistics-viewer.svg'},
+    tags: ['viewer'],
+    name: 'Sequence Mutation Cliffs',
+    description: 'Mutation Cliffs Line Chart',
+    outputs: [{type: 'viewer', name: 'result'}],
+  })
+  static mutationCliffs(): DG.Viewer {
+    return new MutationCliffsViewer();
+  }
+
 
   @grok.decorators.func({
     meta: {icon: 'files/icons/logo-summary-viewer.svg'},

package/src/tests/viewers.ts

@@ -14,24 +14,28 @@ import {TEST_COLUMN_NAMES} from './utils';
 import {showTooltip} from '../utils/tooltips';
 import {PeptideUtils} from '../peptideUtils';
 
-category('Viewers: Basic', () => {
-  let df: DG.DataFrame;
-
-  before(async () => {
-    await PeptideUtils.loadComponents();
-
-    df = DG.DataFrame.fromCsv(await _package.files.readAsText('tests/HELM_small.csv'));
-    await delay(500);
-  });
-
-  const viewers = DG.Func.find({package: 'Peptides', tags: ['viewer']}).map((f) => f.friendlyName);
-  for (const v of viewers) {
-    test(v, async () => {
-      await testViewer(v, df.clone(), {detectSemanticTypes: true, arbitraryDfTest: false, readOnly: true});
-      await delay(1000);
-    });
-  }
-});
+// category('Viewers: Basic', () => {
+//   let df: DG.DataFrame;
+
+//   before(async () => {
+//     await PeptideUtils.loadComponents();
+
+//     df = DG.DataFrame.fromCsv(await _package.files.readAsText('tests/HELM_small.csv'));
+//     const clusterCol = df.col('cluster')!.convertTo(DG.COLUMN_TYPE.STRING);
+//     clusterCol!.name = 'cluster';
+//     df.columns.remove('cluster');
+//     df.columns.add(clusterCol!);
+//     await delay(500);
+//   });
+
+//   const viewers = DG.Func.find({package: 'Peptides', tags: ['viewer']}).map((f) => f.friendlyName);
+//   for (const v of viewers) {
+//     test(v, async () => {
+//       await testViewer(v, df.clone(), {detectSemanticTypes: true, arbitraryDfTest: false, readOnly: true});
+//       await delay(1000);
+//     });
+//   }
+// });
 
 category('Viewers: Monomer-Position', () => {
   let df: DG.DataFrame;

package/src/tests/widgets.ts

@@ -98,7 +98,7 @@ category('Widgets: Distribution panel', () => {
     await delay(1000);
     const lstViewer = model.findViewer(VIEWER_TYPE.LOGO_SUMMARY_TABLE) as LogoSummaryTable | null;
     getDistributionWidget(model.df, {
-      peptideSelection: DG.BitSet.create(model.df.rowCount), columns: model.settings!.columns!,
+      columns: model.settings!.columns!,
       activityCol: scaledActivityCol, clusterSelection: lstViewer!.clusterSelection, clusterColName: clusterCol.name,
       monomerPositionSelection: model.webLogoSelection,
     });

package/src/utils/algorithms.ts

@@ -6,6 +6,7 @@ import {ParallelMutationCliffs} from './parallel-mutation-cliffs';
 import {CLUSTER_TYPE} from '../viewers/logo-summary';
 import BitArray from '@datagrok-libraries/utils/src/bit-array';
 import {
+  bitSetToBitArray,
   ClusterStats,
   ClusterTypeStats,
   getStats,
@@ -18,7 +19,8 @@ import {
 export type MutationCliffsOptions = {
   maxMutations?: number,
   minActivityDelta?: number,
-  filter?: Uint32Array
+  filter?: Uint32Array,
+  singlePosition?: {position: number}
 };
 
 /**
@@ -120,6 +122,7 @@ export function calculateMonomerPositionStatistics(activityCol: DG.Column<number
   //   if (options.aggValue)
   //     options.aggValue.col = options.aggValue.col.clone(filter);
   // }
+  const filterBitArray = options.isFiltered ? bitSetToBitArray(filter) : undefined;
 
   for (const posCol of positionColumns) {
     if (!options.columns.includes(posCol.name))
@@ -144,7 +147,7 @@ export function calculateMonomerPositionStatistics(activityCol: DG.Column<number
       const bitArray = BitArray.fromValues(boolArray);
       if (bitArray.allFalse)
         continue;
-      const stats = getStats(activityColData, bitArray, options.aggValue);
+      const stats = getStats(activityColData, bitArray, filterBitArray, options.aggValue);
       currentPositionObject[monomer] = stats;
       getSummaryStats(currentPositionObject.general, stats);
     }
@@ -245,6 +248,7 @@ export function calculateClusterStatistics(df: DG.DataFrame, clustersColumnName:
   const origClustStats: ClusterStats = {};
   const customClustStats: ClusterStats = {};
 
+  const filterMask = df.filter.anyFalse ? bitSetToBitArray(df.filter) : undefined;
   for (const clustType of Object.values(CLUSTER_TYPE)) {
     const masks = clustType === CLUSTER_TYPE.ORIGINAL ? origClustMasks : customClustMasks;
     const clustNames = clustType === CLUSTER_TYPE.ORIGINAL ? origClustColCat : customClustColNamesList;
@@ -253,7 +257,7 @@ export function calculateClusterStatistics(df: DG.DataFrame, clustersColumnName:
       const mask = masks[maskIdx];
       resultStats[clustNames[maskIdx]] = mask.allTrue || mask.allFalse ?
         {count: mask.length, meanDifference: 0, ratio: 1.0, pValue: null, mask: mask, mean: activityCol.stats.avg} :
-        getStats(activityColData, mask);
+        getStats(activityColData, mask, filterMask);
     }
   }
 

package/src/utils/cell-renderer.ts

@@ -410,7 +410,10 @@ export function setWebLogoRenderer(grid: DG.Grid, monomerPositionStats: MonomerP
         return;
       }
       tooltipOptions.monomerPosition = monomerPosition;
-      requestWebLogoAction(ev, monomerPosition, df, activityCol, options, tooltipOptions);
+      const isDfFiltered = df.filter.anyFalse;
+      const actionDf = isDfFiltered ? df.clone(df.filter) : df;
+      const actionActivityCol = actionDf.getCol(activityCol.name) as DG.Column<number>;
+      requestWebLogoAction(ev, monomerPosition, actionDf, actionActivityCol, options, tooltipOptions);
       if (!options.isSelectionTable && options.highlightCallback != null)
         options.highlightCallback(monomerPosition, df, monomerPositionStats);
     } else if (options.unhighlightCallback != null)

package/src/utils/misc.ts

@@ -131,39 +131,35 @@ export function getDistributionPanel(hist: DG.Viewer<DG.IHistogramSettings>, sta
  * Creates a table to plot activity distribution.
  * @param activityCol - Activity column.
  * @param selection - Selection bitset.
- * @param [peptideSelection] - Peptide selection bitset.
  * @return - Dataframe with activity distribution.
  */
-export function getDistributionTable(activityCol: DG.Column<number>, selection: DG.BitSet, peptideSelection?: DG.BitSet,
-): DG.DataFrame {
-  const selectionMismatch = peptideSelection?.clone().xor(selection).anyTrue ?? false;
+export function getDistributionTable(activityCol: DG.Column<number>, selection: DG.BitSet): DG.DataFrame {
+  if (!activityCol.dataFrame)
+    DG.DataFrame.fromColumns([activityCol]); // to make sure that activityCol has a parent dataframe
+  const filter = activityCol.dataFrame!.filter;
+  const selectionAndFilter = selection.clone().and(filter);
   const rowCount = activityCol.length;
   const activityColData = activityCol.getRawData();
-  const activityData = new Float32Array(rowCount + selection.trueCount +
-    (selectionMismatch ? (peptideSelection?.trueCount ?? 0) : 0));
+  const filteredRowCount = filter.trueCount;
+  const activityData = new Float32Array(filteredRowCount + selectionAndFilter.trueCount);
   const categories: string[] = new Array(activityData.length);
 
-  for (let i = 0, j = 0, k = 0; i < rowCount; ++i) {
-    const isSelected = selection.get(i);
-    activityData[i] = activityColData[i];
-    categories[i] = isSelected ? SPLIT_CATEGORY.SELECTION : SPLIT_CATEGORY.ALL;
+  for (let i = 0, /*J is counting for filtered data*/j = 0, /*selected data*/k = 0; i < rowCount; ++i) {
+    if (!filter.get(i))
+      continue;
+    const isSelected = selectionAndFilter.get(i);
+    activityData[j] = activityColData[i];
+    categories[j] = isSelected ? SPLIT_CATEGORY.SELECTION : SPLIT_CATEGORY.ALL;
+    j++;
     if (isSelected) {
-      activityData[rowCount + j] = activityColData[i];
-      categories[rowCount + j] = SPLIT_CATEGORY.ALL;
-      ++j;
-    }
-    if (selectionMismatch && peptideSelection?.get(i)) {
-      activityData[rowCount + selection.trueCount + k] = activityColData[i];
-      categories[rowCount + selection.trueCount + k] = SPLIT_CATEGORY.PEPTIDES_SELECTION;
+      activityData[filteredRowCount + k] = activityColData[i];
+      categories[filteredRowCount + k] = SPLIT_CATEGORY.ALL;
       ++k;
     }
   }
 
   const splitCol = DG.Column.fromStrings(C.COLUMNS_NAMES.SPLIT_COL, categories);
   const categoryOrder = [SPLIT_CATEGORY.ALL, SPLIT_CATEGORY.SELECTION];
-  if (selectionMismatch)
-    categoryOrder.push(SPLIT_CATEGORY.PEPTIDES_SELECTION);
-
 
   splitCol.setCategoryOrder(categoryOrder);
   splitCol.meta.colors.setCategorical();

package/src/utils/statistics.ts

@@ -1,3 +1,4 @@
+/* eslint-disable max-len */
 import * as DG from 'datagrok-api/dg';
 import {tTest} from '@datagrok-libraries/statistics/src/tests';
 import {RawData} from './types';
@@ -37,24 +38,42 @@ export type SummaryStats = {
 export const getAggregatedColName = (aggF: string, colName: string): string => `${aggF}(${colName})`;
 export type AggregationColumns = { [col: string]: DG.AggregationType };
 
+export function bitSetToBitArray(bitSet: DG.BitSet): BitArray {
+  return BitArray.fromUint32Array(
+    bitSet.length,
+    new Uint32Array(bitSet.getBuffer().buffer as ArrayBuffer),
+  );
+}
+
+export function bitArrayToBitSet(bitArray: BitArray): DG.BitSet {
+  return DG.BitSet.fromBytes(
+    bitArray.buffer.buffer as ArrayBuffer,
+    bitArray.length,
+  );
+}
+
 /**
  * Returns statistics for the given activity data and bit array.
  * @param data - Activity data to calculate statistics for.
  * @param bitArray - Bit array to use for the calculation.
  * @return - Statistics for the given data and bit array.
  */
-export function getStats(data: RawData | number[], bitArray: BitArray,
+export function getStats(data: RawData | number[], bitArray: BitArray, filter?: BitArray,
   aggData?: {col: DG.Column, type: DG.AGG}): StatsItem {
-  if (data.length !== bitArray.length && data.some((v, i) => i >= bitArray.length ? v !== 0 : false))
+  // if column length is less than 16, raw data will be still of length 16
+  if ((data.length !== bitArray.length && data.length !== 16) || (filter && data.length !== filter.length && data.length !== 16))
     throw new Error('PeptidesError: Data and bit array have different lengths');
+  if (filter && filter.allTrue)
+    filter = undefined; // no need to filter if all true
 
-  const selected = new Float32Array(bitArray.trueCount());
-  const rest = new Float32Array(bitArray.falseCount());
+  const filterAndMask = filter ? bitArray.clone().and(filter) : bitArray;
+  const selected = new Float32Array(filterAndMask.trueCount());
+  const rest = new Float32Array(filter ? (filter.trueCount() - filterAndMask.trueCount()) : filterAndMask.falseCount());
   let aggValue: number | undefined;
   if (aggData) {
     try {
       aggValue = DG.DataFrame.fromColumns([aggData.col])
-        .clone(DG.BitSet.fromBytes(bitArray.buffer.buffer, bitArray.length)).col(aggData.col.name)
+        .clone(bitArrayToBitSet(filterAndMask)).col(aggData.col.name)
         ?.aggregate(aggData.type);
     } catch (e) {
       console.error(e);
@@ -63,35 +82,40 @@ export function getStats(data: RawData | number[], bitArray: BitArray,
   let selectedIndex = 0;
   let restIndex = 0;
   for (let i = 0; i < bitArray.length; ++i) {
+    if (filter && !filter.getBit(i))
+      continue;
     if (bitArray.getBit(i))
       selected[selectedIndex++] = data[i];
     else
       rest[restIndex++] = data[i];
   }
+  const totalAnalyzed = selected.length + rest.length;
 
   const selectedMean = selected.reduce((a, b) => a + b, 0) / Math.max(selected.length, 1);
   if (selected.length < 2 || rest.length < 2) {
     const restMean = rest.reduce((a, b) => a + b, 0) / Math.max(rest.length, 1);
+
     return {
       count: selected.length,
       pValue: null,
       mean: selectedMean,
       meanDifference: selectedMean - restMean,
-      ratio: selected.length / (bitArray.length),
-      mask: bitArray,
+      ratio: selected.length / totalAnalyzed,
+      mask: filterAndMask,
       aggValue,
     };
   }
 
   const testResult = tTest(selected, rest);
+
   const currentMeanDiff = testResult['Mean difference']!;
   return {
     count: selected.length,
     pValue: testResult[currentMeanDiff >= 0 ? 'p-value more' : 'p-value less'],
     mean: selectedMean,
     meanDifference: currentMeanDiff,
-    ratio: selected.length / (bitArray.length),
-    mask: bitArray,
+    ratio: selected.length / totalAnalyzed,
+    mask: filterAndMask,
     aggValue,
   };
 }
@@ -129,7 +153,7 @@ export function getAggregatedColumnValues(df: DG.DataFrame, columns: [string, DG
   options.filterDf ??= false;
   options.fractionDigits ??= 3;
 
-  const filteredDf = options.filterDf && df.filter.anyFalse ? df.clone(df.filter) : df;
+  const filteredDf = options.filterDf && df.filter.anyFalse && (!options.mask || options.mask.length === df.filter.trueCount) ? df.clone(df.filter) : df;
 
   const colResults: StringDictionary = {};
   for (const [colName, aggFn] of columns) {

package/src/utils/tooltips.ts

@@ -60,6 +60,7 @@ export function showTooltipAt(df: DG.DataFrame, activityCol: DG.Column<number>,
 
 
     const mask = DG.BitSet.fromBytes(stats.mask.buffer.buffer as ArrayBuffer, activityCol.length);
+    mask.and(df.filter);
     const hist = getActivityDistribution(getDistributionTable(activityCol, mask), true);
     const tableMap = getStatsTableMap(stats);
     if (options.fromViewer) {
@@ -85,6 +86,7 @@ export function showTooltipAt(df: DG.DataFrame, activityCol: DG.Column<number>,
     if (!stats)
       return null;
     const mask = DG.BitSet.fromBytes(stats.mask.buffer.buffer as ArrayBuffer, activityCol.length);
+    mask.and(df.filter);
     const hist = getActivityDistribution(getDistributionTable(activityCol, mask), true);
     const tableMap = getStatsTableMap(stats, {countName: 'Unique count'});
     if (options.fromViewer) {

package/src/viewers/logo-summary.ts

@@ -11,6 +11,7 @@ import * as CR from '../utils/cell-renderer';
 import {HorizontalAlignments, IWebLogoViewer, PositionHeight} from '@datagrok-libraries/bio/src/viewers/web-logo';
 import {
   AggregationColumns,
+  bitSetToBitArray,
   ClusterTypeStats,
   getAggregatedColumnValues,
   getAggregatedValue,
@@ -315,15 +316,30 @@ export class LogoSummaryTable extends DG.JsViewer implements ILogoSummaryTable {
   onTableAttached(): void {
     super.onTableAttached();
     if (isApplicableDataframe(this.dataFrame)) {
-      const activityColName = wu(this.dataFrame.columns.numerical).next().value.name;
-      this.getProperty(`${LST_PROPERTIES.SEQUENCE}${COLUMN_NAME}`)
-        ?.set(this, this.dataFrame.columns.bySemType(DG.SEMTYPE.MACROMOLECULE)!.name);
-      this.getProperty(`${LST_PROPERTIES.ACTIVITY}${COLUMN_NAME}`)
-        ?.set(this, activityColName);
-      this.getProperty(`${LST_PROPERTIES.WEB_LOGO_AGGREGATION_COLUMN_NAME}`)?.set(this, activityColName);
-
-      this.getProperty(`${LST_PROPERTIES.CLUSTERS}${COLUMN_NAME}`)
-        ?.set(this, wu(this.dataFrame.columns.categorical).next().value.name);
+      const potentialActivityColName = wu(this.dataFrame.columns.numerical).filter((col) => col.name?.includes('activ')).take(1).toArray()[0]?.name;
+      const activityColName = potentialActivityColName ??
+        wu(this.dataFrame.columns.numerical).take(1).toArray()?.[0]?.name;
+      const seqCol = this.dataFrame.columns.bySemType(DG.SEMTYPE.MACROMOLECULE);
+      // const clusterCol =
+      if (seqCol) {
+        this.getProperty(`${LST_PROPERTIES.SEQUENCE}${COLUMN_NAME}`)
+          ?.set(this, seqCol.name);
+      }
+      if (activityColName) {
+        this.getProperty(`${LST_PROPERTIES.ACTIVITY}${COLUMN_NAME}`)
+          ?.set(this, activityColName);
+        this.getProperty(`${LST_PROPERTIES.WEB_LOGO_AGGREGATION_COLUMN_NAME}`)?.set(this, activityColName);
+      }
+
+      const potentialClusterCol = wu(this.dataFrame.columns)
+        .filter((col) => (col.isCategorical) && col.name?.toLowerCase().includes('cluster') && col.name != seqCol?.name && col.type !== DG.COLUMN_TYPE.BOOL)?.take(1)?.toArray()?.[0];
+      const clusterCol = potentialClusterCol ??
+        wu(this.dataFrame.columns.categorical)
+          .filter((col) => col.name != seqCol?.name && col.categories.length <= 50 && col.type !== DG.COLUMN_TYPE.BOOL)?.take(1)?.toArray()?.[0];
+      if (clusterCol) {
+        this.getProperty(`${LST_PROPERTIES.CLUSTERS}${COLUMN_NAME}`)
+          ?.set(this, clusterCol.name);
+      }
     } else {
       const msg = 'PeptidesError: dataframe is missing Macromolecule or numeric columns';
       grok.log.error(msg);
@@ -914,8 +930,8 @@ export class LogoSummaryTable extends DG.JsViewer implements ILogoSummaryTable {
     const viewerDfColsLength = viewerDfCols.length;
     const newClusterVals = new Array(viewerDfCols.length);
     const activityScaledCol = this.getScaledActivityColumn();
-    const bitArray = BitArray.fromString(currentSelection.toBinaryString());
-    const stats = getStats(activityScaledCol.getRawData(), bitArray);
+    const bitArray = bitSetToBitArray(currentSelection);
+    const stats = getStats(activityScaledCol.getRawData(), bitArray, bitSetToBitArray(this.dataFrame.filter));
 
     this.bitsets.push(currentSelection.clone());