@platforma-open/milaboratories.top-antibodies.workflow 1.8.2 → 1.10.0

package/.turbo/turbo-build.log

@@ -1,6 +1,6 @@
  WARN  Issue while reading "/home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/.npmrc". Failed to replace env in config: ${NPMJS_TOKEN}
 
-> @platforma-open/milaboratories.top-antibodies.workflow@1.8.2 build /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow
+> @platforma-open/milaboratories.top-antibodies.workflow@1.10.0 build /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow
 > rm -rf dist && pl-tengo check && pl-tengo build
 
 Processing "src/filter-and-sample.tpl.tengo"...

package/CHANGELOG.md

@@ -1,5 +1,27 @@
 # @platforma-open/milaboratories.top-antibodies.workflow
 
+## 1.10.0
+
+### Minor Changes
+
+- c282203: Improved block performance. Fixed increasing ranking order and cluster size ranking
+
+### Patch Changes
+
+- Updated dependencies [c282203]
+  - @platforma-open/milaboratories.top-antibodies.sample-clonotypes@1.3.0
+
+## 1.9.0
+
+### Minor Changes
+
+- b499ab2: Add rank column
+
+### Patch Changes
+
+- Updated dependencies [b499ab2]
+  - @platforma-open/milaboratories.top-antibodies.sample-clonotypes@1.2.0
+
 ## 1.8.2
 
 ### Patch Changes

package/dist/tengo/lib/sampled-cols-conv.lib.tengo

@@ -1,14 +1,7 @@
 ll := import("@platforma-sdk/workflow-tengo:ll")
 
-getColumns := func(datasetSpec) {
-  return {
-    axes: [
-      {
-        column: "clonotypeKey",
-        spec: datasetSpec.axesSpec[1]
-      }],
-    columns: [
-      {
+getColumns := func(datasetSpec, addRanking) {
+  columns := [{
         column: "top",
         id: "link",
         allowNA: false,
@@ -22,8 +15,31 @@ getColumns := func(datasetSpec) {
             "pl7.app/isSubset": "true"
           }
         }
-      }
-    ],
+      }]
+  if addRanking {
+    columns = columns + [{
+        column: "ranked_order",
+        spec: {
+          name: "pl7.app/vdj/ranking-order",
+          valueType: "Int",
+          domain: {},
+          annotations: {
+            "pl7.app/label": "Rank",
+            "pl7.app/table/visibility": "optional",
+            "pl7.app/isSubset": "true"
+          }
+        }
+      }]
+  }
+
+
+  return {
+    axes: [
+      {
+        column: "clonotypeKey",
+        spec: datasetSpec.axesSpec[1]
+      }],
+    columns: columns,
     storageFormat: "Binary",
     partitionKeyLength: 0
   }

package/package.json

@@ -1,11 +1,11 @@
 {
   "name": "@platforma-open/milaboratories.top-antibodies.workflow",
-  "version": "1.8.2",
+  "version": "1.10.0",
   "type": "module",
   "description": "Block Workflow",
   "dependencies": {
     "@platforma-sdk/workflow-tengo": "^4.9.2",
-    "@platforma-open/milaboratories.top-antibodies.sample-clonotypes": "1.1.0",
+    "@platforma-open/milaboratories.top-antibodies.sample-clonotypes": "1.3.0",
     "@platforma-open/milaboratories.top-antibodies.umap": "1.0.3",
     "@platforma-open/milaboratories.top-antibodies.spectratype": "1.4.0"
   },

package/src/filter-and-sample.tpl.tengo

@@ -1,4 +1,4 @@
-// Template for clonotype filtering
+// Template for clonotype filtering and sampling
 self := import("@platforma-sdk/workflow-tengo:tpl")
 exec := import("@platforma-sdk/workflow-tengo:exec")
 assets := import("@platforma-sdk/workflow-tengo:assets")
@@ -15,6 +15,7 @@ self.body(func(inputs) {
     cloneTable := inputs.cloneTable
     datasetSpec := inputs.datasetSpec
     filterMap := inputs.filterMap
+    rankingMap := inputs.rankingMap
     topClonotypes := inputs.topClonotypes
 
     outputs := {}
@@ -39,7 +40,7 @@ self.body(func(inputs) {
     filteredClonotypesCsv := filterResult.getFile("filteredClonotypes.csv")
 
     // Store outputs 
-    sampledColsParams := sampledColsConv.getColumns(datasetSpec)
+    sampledColsParams := sampledColsConv.getColumns(datasetSpec, false) // No ranking column
     filteredClonotypesPf := xsv.importFile(filteredClonotypesCsv, "csv", sampledColsParams,
                                         {cpu: 1, mem: "16GiB"})
 
@@ -58,6 +59,7 @@ self.body(func(inputs) {
 			addFile("filteredClonotypes.csv", filteredClonotypesCsv).
 			arg("--csv").arg("filteredClonotypes.csv").
 			arg("--n").arg(string(topClonotypes)).
+			arg("--ranking-map").arg(string(json.encode(rankingMap))).
 			arg("--out").arg("sampledClonotypes_top.csv").
 			saveFile("sampledClonotypes_top.csv").
 			printErrStreamToStdout().
@@ -69,6 +71,7 @@ self.body(func(inputs) {
 		finalClonotypesCsv = sampleClones.getFile("sampledClonotypes_top.csv")
 		
 		// Store outputs 
+        sampledColsParams := sampledColsConv.getColumns(datasetSpec, true) // Add ranking column
 		sampledColumnsPf := xsv.importFile(finalClonotypesCsv, "csv", sampledColsParams,
 											{cpu: 1, mem: "16GiB"})
 		outputs["sampledRows"] = pframes.exportFrame(sampledColumnsPf)

package/src/main.tpl.tengo

@@ -4,7 +4,6 @@ exec := import("@platforma-sdk/workflow-tengo:exec")
 assets:= import("@platforma-sdk/workflow-tengo:assets")
 xsv := import("@platforma-sdk/workflow-tengo:pframes.xsv")
 pframes := import("@platforma-sdk/workflow-tengo:pframes")
-sampledColsConv := import(":sampled-cols-conv")
 spectratypeConv := import(":pf-spectratype-conv")
 vjUsageConv := import(":pf-vj-usage-conv")
 slices := import("@platforma-sdk/workflow-tengo:slices")
@@ -50,6 +49,12 @@ wf.prepare(func(args){
         partialAxesMatch: true
     }, "linkers")
 
+    // Add cluster size columns from clustering blocks
+    bundleBuilder.addMulti({
+        name: "pl7.app/vdj/clustering/clusterSize",
+        partialAxesMatch: true
+    }, "clusterSizes")
+
     // Add CDR3 sequences
 	bundleBuilder.addMulti({
 		axes: [{ anchor: "main", idx: 1 }], // Clonotype axis
@@ -88,23 +93,24 @@ wf.body(func(args) {
 	// Input arguments
 	columns := args.columns
 	datasetSpec := columns.getSpec(args.inputAnchor)
-	
 	topClonotypes := args.topClonotypes
 
-	// Needed conditional variables
+	// Needed conditional variable
 	isSingleCell := datasetSpec.axesSpec[1].name == "pl7.app/vdj/scClonotypeKey"
 	
-	// output containers 
+	// Output container
 	outputs := {}
 
-	// Get filtered clonotypes from prerun
-	// Build clonotype table
+	// Build clonotype table csv for filtering script
 	cloneTable := pframes.csvFileBuilder() 
 	cloneTable.setAxisHeader(datasetSpec.axesSpec[1].name, "clonotypeKey")
 
-	// Add Filters to table
-	addedAxes := []
-	filterMap := {}
+	// Add filter columns to table
+	// Filter structure {id: UI id, value: AnchoredColumnId, filter: Filter criteria, isExpanded: boolean (UI state)}
+	// AnchoredColumnId {anchorRef: PlRef, anchorName: string, column: SUniversalPColumnId (unique column id)}
+	addedAxes := [] // Keep track of axes that are added to the table to prevent duplicates
+	filterMap := {} // Map column headers to filter criteria
+	rankingMap := {} // Map column headers to ranking order (increasing/decreasing)
 	if len(args.filters) > 0 {
 		for i, filter in args.filters {
 			if filter.value != undefined {
@@ -114,9 +120,9 @@ wf.body(func(args) {
 				// Store reference value and filter type associated to this column
 				filterMap["Filter_" + string(i)] = filter.filter
 			
-				// If column does not have main anchor axis we have to include theirs
+				// If column does not have main anchor axis we have to include theirs (columns coming from clustering blocks for example)
 				colsSpec := columns.getSpec(filter.value.column)
-				axesNames := slices.map(colsSpec.axesSpec, func (a) { return a.name})
+				axesNames := slices.map(colsSpec.axesSpec, func (a) {return a.name})
 				if !slices.hasElement(axesNames, datasetSpec.axesSpec[1].name) {
 					for na, ax in colsSpec.axesSpec {
 						if ax.name != datasetSpec.axesSpec[1].name {
@@ -133,13 +139,15 @@ wf.body(func(args) {
 	if len(args.rankingOrder) > 0 {
 		for i, col in args.rankingOrder {
 			cloneTable.add(columns.getColumn(col.value.column), {header: "Col" + string(i)})
+			// Store ranking order for this column
+			rankingMap["Col" + string(i)] = col.rankingOrder
 		
-			// If column does not have main anchor axis we have to include theirs
+			// If column does not have main anchor axis we have to include theirs (columns coming from clustering blocks for example)
 			colsSpec := columns.getSpec(col.value.column)
-			axesNames := slices.map(colsSpec.axesSpec, func (a) { return a.name})
+			axesNames := slices.map(colsSpec.axesSpec, func (a) {return a.name})
 			if !slices.hasElement(axesNames, datasetSpec.axesSpec[1].name) {
 				for na, ax in colsSpec.axesSpec {
-					if ax.name != datasetSpec.axesSpec[1].name && !slices.hasElement(addedAxes, ax.name) {
+					if ax.name != datasetSpec.axesSpec[1].name && !slices.hasElement(addedAxes, ax.name) { // Prevent duplicates
 						cloneTable.setAxisHeader(ax.name, "cluster_" + string(i) + string(na))
 					}
 				}
@@ -148,6 +156,8 @@ wf.body(func(args) {
 	} else {
 		i := 0
 		cloneTable.add(columns.getColumn(args.rankingOrderDefault.value.column), {header: "Col" + string(i)})
+		// Store default ranking order
+		rankingMap["Col" + string(i)] = args.rankingOrderDefault.rankingOrder
 	
 		// If column does not have main anchor axis we have to include theirs
 		colsSpec := columns.getSpec(args.rankingOrderDefault.value.column)
@@ -161,8 +171,8 @@ wf.body(func(args) {
 		}	
 	}
 
-	// Get linker columns if needed
-	linkerAxisSpec := {}
+	// Add linker columns when needed
+	linkerAxisSpec := {} // Map cluster axis names to specs. Is it needed????
 	if len(columns.getColumns("linkers")) > 0 {
 		for i, col in columns.getColumns("linkers") {
 			if datasetSpec.axesSpec[1].name == col.spec.axesSpec[1].name {
@@ -176,25 +186,40 @@ wf.body(func(args) {
 			}
 		}
 	}
+
+	// Add cluster size columns if available
+	if len(columns.getColumns("clusterSizes")) > 0 {
+		for i, col in columns.getColumns("clusterSizes") {
+			cloneTable.add(col, {header: "clusterSize." + string(i)})
+			// Add the cluster axis header
+			for axisIdx, axis in col.spec.axesSpec {
+				if axis.name != datasetSpec.axesSpec[1].name {
+					cloneTable.setAxisHeader(axis.name, "clusterAxis_" + string(i) + "_" + string(axisIdx))
+				}
+			}
+		}
+	}
+
 	cloneTable.mem("16GiB")
 	cloneTable.cpu(1)
 	cloneTable = cloneTable.build()
 
-	// Use ender.create to call the filter-clonotypes template
-	filterResult := render.create(filterAndSampleTpl, {
+	// Use render.create to call the filtering and sampling clonotypes template
+	filterSampleResult := render.create(filterAndSampleTpl, {
 		inputAnchor: args.inputAnchor,
 		cloneTable: cloneTable,
 		rankingOrder: args.rankingOrder,
 		rankingOrderDefault: args.rankingOrderDefault,
 		filters: args.filters,
 		filterMap: filterMap,
+		rankingMap: rankingMap,
 		datasetSpec: datasetSpec,
 		topClonotypes: args.topClonotypes
 	})
 	
-	// Get the filtered clonotypes P-frame and CSV from the template result
-	finalClonotypesCsv := filterResult.output("finalClonotypesCsv", 24 * 60 * 60 * 1000)
-	// outputs["sampledRows"] = filterResult.output("sampledRows", 24 * 60 * 60 * 1000)
+	// Get the filtered and sampled clonotypes P-frame and CSV from the template result
+	finalClonotypesCsv := filterSampleResult.output("finalClonotypesCsv", 24 * 60 * 60 * 1000)
+	// outputs["sampledRows"] = filterSampleResult.output("sampledRows", 24 * 60 * 60 * 1000)
 	
 	////////// CDR3 Length Calculation //////////
 	
@@ -217,15 +242,15 @@ wf.body(func(args) {
 			chain := col.spec.domain["pl7.app/vdj/scClonotypeChain"]  // e.g., "A", "B"
 			receptor := col.spec.axesSpec[0].domain["pl7.app/vdj/receptor"]  // e.g., "IG", "TCRAB", "TCRGD"
 			chainLabel := chainMapping[receptor][chain]
-			return baseHeaderName + "." + chainLabel
+			return baseHeaderName + "." + chainLabel // e.g., "cdr3Sequence.Heavy"
 		} else {
 			// For bulk, if chain info is available (e.g. IGH, IGK, IGL)
 			chainFromDomain := col.spec.axesSpec[0].domain["pl7.app/vdj/chain"] // e.g. "IGH", "IGK"
 			if chainFromDomain != undefined {
-				return baseHeaderName + "." + chainFromDomain
+				return baseHeaderName + "." + chainFromDomain // e.g., "cdr3Sequence.IGH"
 			}
 		}
-		return baseHeaderName // Default header for bulk
+		return baseHeaderName
 	};
 
 	// Process CDR3 sequences
@@ -251,6 +276,7 @@ wf.body(func(args) {
 		headerName := makeHeaderName(col, "jGene", isSingleCell)
 		cdr3SeqTable.add(col, {header: headerName})
 	}
+
 	cdr3SeqTable.mem("16GiB")
 	cdr3SeqTable.cpu(1)
 	cdr3SeqTableBuilt := cdr3SeqTable.build()
@@ -280,7 +306,7 @@ wf.body(func(args) {
 		run()
 
 
-	// For spectratype structure is [chain][cdr3Length][vGene] -> count
+	// Spectratype PFrame structure is [chain][cdr3Length][vGene] -> count
 
 	cdr3VspectratypePf := xsv.importFile(cdr3VspectratypeCmd.getFile("spectratype.tsv"), 
 										"tsv", spectratypeConv.getColumns(),

package/src/prerun.tpl.tengo

@@ -51,6 +51,12 @@ wf.prepare(func(args){
         partialAxesMatch: true
     }, "linkers")
 
+    // Add cluster size columns from clustering blocks
+    bundleBuilder.addMulti({
+        name: "pl7.app/vdj/clustering/clusterSize",
+        partialAxesMatch: true
+    }, "clusterSizes")
+
     // Add CDR3 sequences
 	bundleBuilder.addMulti({
 		axes: [{ anchor: "main", idx: 1 }], // Clonotype axis
@@ -100,6 +106,7 @@ wf.body(func(args) {
         // Add Filters to table
         addedAxes := []
         filterMap := {}
+        rankingMap := {}
         if len(args.filters) > 0 {
             for i, filter in args.filters {
                 if filter.value != undefined {
@@ -129,6 +136,8 @@ wf.body(func(args) {
             for i, col in args.rankingOrder {
                 if col.value != undefined {
                     cloneTable.add(columns.getColumn(col.value.column), {header: "Col" + string(i)})
+                    // Store ranking order for this column
+                    rankingMap["Col" + string(i)] = col.rankingOrder
                 
                     // If column does not have main anchor axis we have to include theirs
                     colsSpec := columns.getSpec(col.value.column)
@@ -148,6 +157,8 @@ wf.body(func(args) {
             if args.rankingOrderDefault.value != undefined {
                 i := 0
                 cloneTable.add(columns.getColumn(args.rankingOrderDefault.value.column), {header: "Col" + string(i)})
+                // Store default ranking order
+                rankingMap["Col" + string(i)] = args.rankingOrderDefault.rankingOrder
             
                 // If column does not have main anchor axis we have to include theirs
                 colsSpec := columns.getSpec(args.rankingOrderDefault.value.column)
@@ -177,12 +188,26 @@ wf.body(func(args) {
                 }
             }
         }
+
+        // Add cluster size columns if available
+        if len(columns.getColumns("clusterSizes")) > 0 {
+            for i, col in columns.getColumns("clusterSizes") {
+                cloneTable.add(col, {header: "clusterSize." + string(i)})
+                // Add the cluster axis header
+                for axisIdx, axis in col.spec.axesSpec {
+                    if axis.name != datasetSpec.axesSpec[1].name {
+                        cloneTable.setAxisHeader(axis.name, "clusterAxis_" + string(i) + "_" + string(axisIdx))
+                    }
+                }
+            }
+        }
+
         cloneTable.mem("16GiB")
         cloneTable.cpu(1)
         cloneTable = cloneTable.build()
 
         // Use ender.create to call the filter-clonotypes template
-        filterResult := render.create(filterAndSampleTpl, {
+        filterSampleResult := render.create(filterAndSampleTpl, {
             inputAnchor: args.inputAnchor,
             cloneTable: cloneTable,
             topClonotypes: args.topClonotypes,
@@ -190,11 +215,12 @@ wf.body(func(args) {
             rankingOrderDefault: args.rankingOrderDefault,
             filters: args.filters,
             filterMap: filterMap,
+            rankingMap: rankingMap,
             datasetSpec: datasetSpec
         })
         
         // Get the filtered clonotypes from the template result
-        outputs["sampledRows"] = filterResult.output("sampledRows", 24 * 60 * 60 * 1000)
+        outputs["sampledRows"] = filterSampleResult.output("sampledRows", 24 * 60 * 60 * 1000)
     }
 
 	return {

package/src/sampled-cols-conv.lib.tengo

@@ -1,14 +1,7 @@
 ll := import("@platforma-sdk/workflow-tengo:ll")
 
-getColumns := func(datasetSpec) {
-  return {
-    axes: [
-      {
-        column: "clonotypeKey",
-        spec: datasetSpec.axesSpec[1]
-      }],
-    columns: [
-      {
+getColumns := func(datasetSpec, addRanking) {
+  columns := [{
         column: "top",
         id: "link",
         allowNA: false,
@@ -22,8 +15,31 @@ getColumns := func(datasetSpec) {
             "pl7.app/isSubset": "true"
           }
         }
-      }
-    ],
+      }]
+  if addRanking {
+    columns = columns + [{
+        column: "ranked_order",
+        spec: {
+          name: "pl7.app/vdj/ranking-order",
+          valueType: "Int",
+          domain: {},
+          annotations: {
+            "pl7.app/label": "Rank",
+            "pl7.app/table/visibility": "optional",
+            "pl7.app/isSubset": "true"
+          }
+        }
+      }]
+  }
+
+
+  return {
+    axes: [
+      {
+        column: "clonotypeKey",
+        spec: datasetSpec.axesSpec[1]
+      }],
+    columns: columns,
     storageFormat: "Binary",
     partitionKeyLength: 0
   }