Optimization to how groupsamples are created; by using the sample unique id we can create sampleidentifier

Author: i-am-leslie

src/pages/studyView/StudyViewPageStore.ts

@@ -1650,11 +1650,12 @@ export class StudyViewPageStore
                     chartMeta.uniqueKey
                 );
                 if (!chartInfo || !this.hasFilteredSamples) return;
+                const includeSampleIds = true;
 
                 const data: any[] = await invokeGenomicDataCountIncludeSampleid(
                     chartInfo,
                     this.filters,
-                    true
+                    includeSampleIds
                 );
 
                 const lcValueToColor = _.keyBy(clinicalAttributeValues, d =>
@@ -1672,7 +1673,7 @@ export class StudyViewPageStore
 
                 /**
                  * Step 2:
-                 * Build SessionGroupData objects for each group, using the sample ids from step 1
+                 * Build SessionGroupData objects for each group, using the sample identifiers from step 1
                  */
                 const groups: SessionGroupData[] = _.chain(
                     clinicalAttributeValues
@@ -1697,9 +1698,9 @@ export class StudyViewPageStore
                             : undefined
                     )
                     .value();
-
                 statusCallback(LoadingPhase.CREATING_SESSION);
 
+                // create session and get id
                 const { id } = await comparisonClient.addComparisonSession({
                     groups,
                     clinicalAttributeName: chartMeta.displayName,

src/pages/studyView/StudyViewUtils.spec.tsx

@@ -3671,11 +3671,7 @@ describe('StudyViewUtils', () => {
                 counts: [
                     {
                         value: 'Missense_Mutation',
-                        sampleIds: [
-                            'study_ABC_123',
-                            'study_DEF_456',
-                            'lgg_ucsf_2014_P21_Pri',
-                        ],
+                        sampleIds: ['study_ABC_123', 'study_DEF_456'],
                     },
                 ],
             },
@@ -3689,12 +3685,7 @@ describe('StudyViewUtils', () => {
             },
         ] as any;
 
-        const studyId: string[] = [
-            'study_ABC',
-            'study_DEF',
-            'study_GHI',
-            'lgg_ucsf_2014',
-        ];
+        const studyId: string[] = ['study_ABC', 'study_DEF', 'study_GHI'];
 
         it('groups and creates SampleIdeifier  by mutation status correctly', () => {
             const result = groupSamplesByMutationStatus(data, studyId);
@@ -3708,10 +3699,6 @@ describe('StudyViewUtils', () => {
                         studyId: 'study_DEF',
                         sampleId: '456',
                     },
-                    {
-                        studyId: 'lgg_ucsf_2014',
-                        sampleId: 'P21_Pri',
-                    },
                 ],
                 NOT_MUTATED: [
                     {

src/pages/studyView/StudyViewUtils.tsx

@@ -4286,20 +4286,21 @@ export function groupSamplesByMutationStatus(data: any[], studyIds: string[]) {
 
     return _.chain(data)
         .flatMap(d => d.counts ?? [])
-        .filter(c => !SKIP_VALUES.has(c.value) && c.sampleIds?.length > 0)
+        .filter(c => !SKIP_VALUES.has(c.value) && c.sampleIds?.length > 0) // avoid NOT_PROFILED samples and any bins without samples
         .flatMap(c => {
             const group = NON_MUTATION_VALUES.has(c.value)
                 ? c.value
                 : 'MUTATED';
 
             return c.sampleIds
                 .map((rawId: string) => {
+                    // sampleIds are in the format of {studyId}_{sampleId}, need to be split before grouping
                     const matchedStudyId = studyIds.find(studyId =>
                         rawId.startsWith(studyId + '_')
                     );
 
                     if (!matchedStudyId) return null;
-
+                    // create an object with group, studyId and sampleId for each sampleId, where group is MUTATED or NOT_MUTATED based on the value of the bin it belongs to
                     return {
                         group,
                         studyId: matchedStudyId,
@@ -4311,8 +4312,14 @@ export function groupSamplesByMutationStatus(data: any[], studyIds: string[]) {
         .groupBy('group')
         .mapValues(items =>
             _.uniqBy(
-                items.map(({ studyId, sampleId }) => ({ studyId, sampleId })),
-                i => i.sampleId
+                // cretaes a sampleidentifier{sampleId, studyId} using the matched studyId and the rest of the rawId as sampleId
+                items.map(
+                    ({ studyId, sampleId }): SampleIdentifier => ({
+                        studyId,
+                        sampleId,
+                    })
+                ),
+                i => i.sampleId // if sample IDs aren't globally unique across studies, we may want to use both studyId and sampleId to determine uniqueness
             )
         )
         .value();