@platforma-open/milaboratories.top-antibodies.workflow 1.2.0 → 1.3.0

package/.turbo/turbo-build.log

@@ -1,15 +1,19 @@
  WARN  Issue while reading "/home/runner/work/top-antibodies/top-antibodies/.npmrc". Failed to replace env in config: ${NPMJS_TOKEN}
 
-> @platforma-open/milaboratories.top-antibodies.workflow@1.2.0 build /home/runner/work/top-antibodies/top-antibodies/workflow
+> @platforma-open/milaboratories.top-antibodies.workflow@1.3.0 build /home/runner/work/top-antibodies/top-antibodies/workflow
 > rm -rf dist && pl-tengo check && pl-tengo build
 
 Processing "src/main.tpl.tengo"...
+Processing "src/pf-spectratype-conv.lib.tengo"...
 Processing "src/pf-umap-conv.lib.tengo"...
+Processing "src/pf-vj-usage-conv.lib.tengo"...
 Processing "src/sampled-cols-conv.lib.tengo"...
 Processing "src/sampled-cols-umap-conv.lib.tengo"...
 No syntax errors found.
   info: Compiling 'dist'...
+  info:   - writing /home/runner/work/top-antibodies/top-antibodies/workflow/dist/tengo/lib/pf-spectratype-conv.lib.tengo
   info:   - writing /home/runner/work/top-antibodies/top-antibodies/workflow/dist/tengo/lib/pf-umap-conv.lib.tengo
+  info:   - writing /home/runner/work/top-antibodies/top-antibodies/workflow/dist/tengo/lib/pf-vj-usage-conv.lib.tengo
   info:   - writing /home/runner/work/top-antibodies/top-antibodies/workflow/dist/tengo/lib/sampled-cols-conv.lib.tengo
   info:   - writing /home/runner/work/top-antibodies/top-antibodies/workflow/dist/tengo/lib/sampled-cols-umap-conv.lib.tengo
   info:   - writing /home/runner/work/top-antibodies/top-antibodies/workflow/dist/tengo/tpl/main.plj.gz

package/CHANGELOG.md

@@ -1,5 +1,23 @@
 # @platforma-open/milaboratories.top-antibodies.workflow
 
+## 1.3.0
+
+### Minor Changes
+
+- 2e24f7a: Disable default normalization in VJ usage plot and change spectratype/VJ usage script to run on top clonotypes if provided
+
+### Patch Changes
+
+- Updated dependencies [2e24f7a]
+  - @platforma-open/milaboratories.top-antibodies.spectratype@1.3.0
+
+## 1.2.1
+
+### Patch Changes
+
+- Updated dependencies [6443da1]
+  - @platforma-open/milaboratories.top-antibodies.spectratype@1.2.0
+
 ## 1.2.0
 
 ### Minor Changes

package/dist/tengo/lib/pf-spectratype-conv.lib.tengo

@@ -0,0 +1,59 @@
+ll := import("@platforma-sdk/workflow-tengo:ll")
+
+getColumns := func() {
+    return {
+        "axes": [
+            {
+                "column": "chain",
+                "spec": {
+                    "name": "pl7.app/vdj/chain",
+                    "type": "String",
+                    "annotations": { "pl7.app/label": "CDR3 chain" }
+                }
+            },
+            {
+                "column": "cdr3Length",
+                "spec": {
+                    "name": "pl7.app/vdj/sequenceLength",
+                    "type": "Int",
+                    "domain": {
+                        "pl7.app/vdj/feature": "CDR3",
+                        "pl7.app/alphabet": "aminoacid"
+                    },
+                    "annotations": { "pl7.app/label": "CDR3 aa Length" }
+                }
+            },
+            {
+                "column": "vGene",
+                "spec": {
+                    "name": "pl7.app/vdj/geneHit",
+                    "type": "String",
+                    "domain": { "pl7.app/vdj/reference": "VGene" },
+                    "annotations": {
+                        "pl7.app/label": "Best V gene"
+                    }
+                }
+            }
+        ],
+        "columns": [
+            {
+                "column": "count",
+                "spec": {
+                    "name": "pl7.app/vdj/vSpectratype",
+                    "valueType": "Int",
+                    "domain": {
+                        "pl7.app/vdj/feature": "CDR3",
+                        "pl7.app/alphabet": "aminoacid"
+                    },
+                    "annotations": { "pl7.app/label": "CDR3 V Spectratype" }
+                }
+            }
+        ],
+        "storageFormat": "Binary",
+        "partitionKeyLength": 0
+    }
+}
+
+export ll.toStrict({
+    getColumns: getColumns
+})

package/dist/tengo/lib/pf-vj-usage-conv.lib.tengo

@@ -0,0 +1,54 @@
+ll := import("@platforma-sdk/workflow-tengo:ll")
+
+getColumns := func() {
+    return {
+        "axes": [
+                {
+                "column": "chain",
+                "spec": {
+                    "name": "pl7.app/vdj/chain",
+                    "type": "String",
+                    "annotations": { "pl7.app/label": "CDR3 chain" }
+                }
+            },
+            {
+                "column": "vGene",
+                "spec": {
+                    "name": "pl7.app/vdj/geneHit",
+                    "type": "String",
+                    "domain": { "pl7.app/vdj/reference": "VGene" },
+                    "annotations": {
+                        "pl7.app/label": "Best V gene"
+                    }
+                }
+            },
+            {
+                "column": "jGene",
+                "spec": {
+                    "name": "pl7.app/vdj/geneHit",
+                    "type": "String",
+                    "domain": { "pl7.app/vdj/reference": "JGene" },
+                    "annotations": {
+                        "pl7.app/label": "Best J gene"
+                    }
+                }
+            }
+        ],
+        "columns": [
+            {
+                "column": "count",
+                "spec": {
+                    "name": "pl7.app/vdj/vjGeneUsage",
+                    "valueType": "Int",
+                    "annotations": { "pl7.app/label": "V/J usage" }
+                }
+            }
+        ],
+        "storageFormat": "Binary",
+        "partitionKeyLength": 0
+    }
+}
+
+export ll.toStrict({
+    getColumns: getColumns
+})

package/package.json

@@ -1,17 +1,17 @@
 {
   "name": "@platforma-open/milaboratories.top-antibodies.workflow",
-  "version": "1.2.0",
+  "version": "1.3.0",
   "type": "module",
   "description": "Block Workflow",
   "dependencies": {
-    "@platforma-sdk/workflow-tengo": "^4.3.2",
+    "@platforma-sdk/workflow-tengo": "^4.6.1",
+    "@platforma-open/milaboratories.top-antibodies.spectratype": "1.3.0",
     "@platforma-open/milaboratories.top-antibodies.sample-clonotypes": "1.0.1",
-    "@platforma-open/milaboratories.top-antibodies.spectratype": "1.1.0",
     "@platforma-open/milaboratories.top-antibodies.umap": "1.0.1"
   },
   "devDependencies": {
-    "@platforma-sdk/tengo-builder": "^2.1.3",
-    "@platforma-sdk/test": "^1.30.24",
+    "@platforma-sdk/tengo-builder": "^2.1.5",
+    "@platforma-sdk/test": "^1.31.16",
     "vitest": "^2.1.8"
   },
   "scripts": {

package/src/main.tpl.tengo

@@ -6,6 +6,8 @@ xsv := import("@platforma-sdk/workflow-tengo:pframes.xsv")
 pframes := import("@platforma-sdk/workflow-tengo:pframes")
 sampledColsConv := import(":sampled-cols-conv")
 sampledColsUmapConv := import(":sampled-cols-umap-conv")
+spectratypeConv := import(":pf-spectratype-conv")
+vjUsageConv := import(":pf-vj-usage-conv")
 slices := import("@platforma-sdk/workflow-tengo:slices")
 umapConv := import(":pf-umap-conv")
 
@@ -17,7 +19,8 @@ wf.prepare(func(args){
 	bundleBuilder.addAnchor("main", args.inputAnchor) 
 	
 	for col in args.rankingOrder {
-		bundleBuilder.addSingle(col)
+		bundleBuilder.addAnchor(col.value.anchorName, col.value.anchorRef)
+		bundleBuilder.addSingle(col.value.column)
 	}
 
 	// Add linker column
@@ -58,6 +61,15 @@ wf.prepare(func(args){
 		}
 	}, "VGenes")
 
+	// Add J gene
+	bundleBuilder.addMulti({
+		axes: [{ anchor: "main", idx: 1 }], // Clonotype axis
+		name: "pl7.app/vdj/geneHit",
+		domain: {
+			"pl7.app/vdj/reference": "JGene"
+		}
+	}, "JGenes")
+
 	return {
 		columns: bundleBuilder.build()
 	}
@@ -80,14 +92,14 @@ wf.body(func(args) {
 	////////// Clonotype Filtering //////////
 	
 	// Build clonotype table
-	cloneTable := columns.xsvTableBuilder()
+	cloneTable := pframes.csvFileBuilder() 
 	cloneTable.setAxisHeader(datasetSpec.axesSpec[1].name, "clonotypeKey")
 	
 	for i, col in args.rankingOrder {
-		cloneTable.add(col, {header: "Col" + string(i)})
+		cloneTable.add(columns.getColumn(col.value.column), {header: "Col" + string(i)})
 	
 		// If column does not have main anchor axis we have to include theirs
-		colsSpec := columns.getSpec(col)
+		colsSpec := columns.getSpec(col.value.column)
 		axesNames := slices.map(colsSpec.axesSpec, func (a) { return a.name})
 		if !slices.hasElement(axesNames, datasetSpec.axesSpec[1].name) {
 			for na, ax in colsSpec.axesSpec {
@@ -102,16 +114,18 @@ wf.body(func(args) {
 	linkerAxisSpec := {}
 	for i, col in columns.getColumns("linkers") {
 		if datasetSpec.axesSpec[1].name == col.spec.axesSpec[1].name {
-			cloneTable.add(col.key, {header: "linker." + string(i)})
+			cloneTable.add(col, {header: "linker." + string(i)})
 			cloneTable.setAxisHeader(col.spec.axesSpec[0].name, "cluster_" + string(i))
 			linkerAxisSpec["cluster_" + string(i)] = col.spec.axesSpec[0]
 		} else if datasetSpec.axesSpec[1].name == col.spec.axesSpec[0].name {
-			cloneTable.add(col.key, {header: "linker." + string(i)})
+			cloneTable.add(col, {header: "linker." + string(i)})
 			cloneTable.setAxisHeader(col.spec.axesSpec[1].name, "cluster_" + string(i))
 			linkerAxisSpec["cluster_" + string(i)] = col.spec.axesSpec[1]
 		}
 	}
-	cloneTable = cloneTable.build("csv")
+	cloneTable = cloneTable.build()
+
+	topClonotypesCsv := undefined
 
 	if topClonotypes != undefined {
 		// Run sampling script
@@ -126,33 +140,36 @@ wf.body(func(args) {
 			saveStdoutContent().
 			cache(24 * 60 * 60 * 1000).
 			run()
-		
+
+		// Save top clonotypes CSV file
+		topClonotypesCsv = sampleClones.getFile("sampledClonotypes_top.csv")
+
 		// Store outputs
 		sampledColsParams := sampledColsConv.getColumns(datasetSpec, linkerAxisSpec)
-		sampledColumnsPf := xsv.importFile(sampleClones.getFile("sampledClonotypes_top.csv"), "csv", sampledColsParams)
+		sampledColumnsPf := xsv.importFile(topClonotypesCsv, "csv", sampledColsParams)
 		outputs["sampledRows"] = pframes.exportFrame(sampledColumnsPf)
 
     	// Prepare filter col subset for UMAP
     	// Avoid taking cluster axis using other params
     	sampledColsUmapParams := sampledColsUmapConv.getColumns(datasetSpec) 
-    	sampledColsUmapPf := xsv.importFile(sampleClones.getFile("sampledClonotypes_top.csv"), "csv", sampledColsUmapParams)
+    	sampledColsUmapPf := xsv.importFile(topClonotypesCsv, "csv", sampledColsUmapParams)
 		outputs["sampledRowsUmap"] = pframes.exportFrame(sampledColsUmapPf)
 	}
 	
 	////////// UMAP //////////
 	// Generate input TSV with Clonotype ID and aa sequence
-	umapTable := columns.xsvTableBuilder()
+	umapTable := pframes.tsvFileBuilder()
 	umapTable.setAxisHeader(datasetSpec.axesSpec[1].name, "clonotypeKey")
 	for col in columns.getColumns("aaSequence") {
 		if isSingleCell {
 			chainLabel := col.spec.domain["pl7.app/vdj/scClonotypeChain"]
-			umapTable.add(col.key, {header: "aaSequence." + chainLabel})
+			umapTable.add(col, {header: "aaSequence." + chainLabel})
 		} else {
 			chainLabel := col.spec.axesSpec[0].domain["pl7.app/vdj/chain"]
-			umapTable.add(col.key, {header: "aaSequence." + chainLabel})
+			umapTable.add(col, {header: "aaSequence." + chainLabel})
 		}
 	}
-	umapTable = umapTable.build("tsv")
+	umapTable = umapTable.build()
 
 	// UMAP script should go here
 	umapClones := exec.builder().
@@ -172,7 +189,7 @@ wf.body(func(args) {
 
 	////////// CDR3 Length Calculation //////////
 	
-	cdr3SeqTable := columns.xsvTableBuilder()
+	cdr3SeqTable := pframes.tsvFileBuilder()
 	cdr3SeqTable.setAxisHeader(datasetSpec.axesSpec[1].name, "clonotypeKey")
 
     // Must deal with multiple CDR3 sequences (two for each cell in single cell data)
@@ -207,7 +224,7 @@ wf.body(func(args) {
 
 	for col in cdr3Sequences {
 		headerName := makeHeaderName(col, "cdr3Sequence", isSingleCell)
-		cdr3SeqTable.add(col.key, {header: headerName})
+		cdr3SeqTable.add(col, {header: headerName})
 	}
 
 	// Process V genes
@@ -215,92 +232,50 @@ wf.body(func(args) {
 
 	for col in vGenes {
 		headerName := makeHeaderName(col, "vGene", isSingleCell)
-		cdr3SeqTable.add(col.key, {header: headerName})
+		cdr3SeqTable.add(col, {header: headerName})
+	}
+
+	// Process J genes
+	jGenes := columns.getColumns("JGenes")	
+
+	for col in jGenes {
+		headerName := makeHeaderName(col, "jGene", isSingleCell)
+		cdr3SeqTable.add(col, {header: headerName})
 	}
 
-	cdr3SeqTableBuilt := cdr3SeqTable.build("tsv")
+	cdr3SeqTableBuilt := cdr3SeqTable.build()
 
 	cdr3VspectratypeCmd := exec.builder().
 		software(assets.importSoftware("@platforma-open/milaboratories.top-antibodies.spectratype:main")).
 		addFile("cdr3_sequences_input.tsv", cdr3SeqTableBuilt).
 		arg("--input_tsv").arg("cdr3_sequences_input.tsv").
-		arg("--output_tsv").arg("cdr3_lengths.tsv").
-		saveFile("cdr3_lengths.tsv").
+		arg("--spectratype_tsv").arg("spectratype.tsv").
+		arg("--vj_usage_tsv").arg("vj_usage.tsv") // no dot here
+
+	// Add top clonotypes argument and file to the builder if provided
+	if topClonotypes != undefined {
+		cdr3VspectratypeCmd = cdr3VspectratypeCmd.
+			arg("--top_clonotypes_csv").arg("topClonotypes.csv").
+			addFile("topClonotypes.csv", topClonotypesCsv)
+	}
+
+	cdr3VspectratypeCmd = cdr3VspectratypeCmd. // continue building the command
+		saveFile("spectratype.tsv").
+		saveFile("vj_usage.tsv").
 		printErrStreamToStdout().
 		saveStdoutContent().
 		cache(24 * 60 * 60 * 1000).
 		run()
 
 
-	// For spectratype structure is:
-	// [chain][cdr3Length][vGene] -> count
+	// For spectratype structure is [chain][cdr3Length][vGene] -> count
 
-	// Get the spec for the vGene column
-	vGeneSpec := columns.getSpec(vGenes[0].key)
-
-	// Set up the axes for the output data
-	axes := [
-		{
-			column: "chain",
-			spec: {
-				name: "pl7.app/vdj/chain",
-				type: "String", // For axis it is type, not valueType 
-				annotations: { "pl7.app/label": "CDR3 chain" }
-			}
-		},
-				{
-			column: "cdr3Length",
-			spec: {
-				name: "pl7.app/vdj/sequenceLength",
-				type: "Int",
-				domain: {
-					"pl7.app/vdj/feature": "CDR3",
-					"pl7.app/alphabet": "aminoacid"
-				},
-				annotations: { "pl7.app/label": "CDR3 aa Length" }
-			}
-		},
-		{
-			column: "vGene",
-			spec: {
-				name: "pl7.app/vdj/geneHit",
-				type: "String",
-				domain: vGeneSpec.domain,
-				annotations: {
-					"pl7.app/label": "Best V gene"
-				}
-			}
-		}
-		
-	]
-
-
-	spectratypeColumns := [
-		{
-			column: "count",
-			spec: {
-				name: "pl7.app/vdj/vSpectratype",
-				valueType: "Int",
-				domain: {
-					"pl7.app/vdj/feature": "CDR3",
-					"pl7.app/alphabet": "aminoacid"
-				},
-				annotations: { "pl7.app/label": "CDR3 V Spectratype" }
-			}
-		}
-	]
-
-	spectratypeSpec := {
-		axes: axes,
-		columns: spectratypeColumns,
-		storageFormat: "Binary",
-		partitionKeyLength: 0
-	}
-
-	cdr3VspectratypePf := xsv.importFile(cdr3VspectratypeCmd.getFile("cdr3_lengths.tsv"), "tsv", spectratypeSpec)
-	
+	cdr3VspectratypePf := xsv.importFile(cdr3VspectratypeCmd.getFile("spectratype.tsv"), "tsv", spectratypeConv.getColumns())
 	outputs["cdr3VspectratypePf"] = pframes.exportFrame(cdr3VspectratypePf) 
-	//ll.print("CDR3 lengths PFrame imported.")
+
+	// For vjUsage structure is [chain][vGene][jGene] -> count
+	vjUsagePf := xsv.importFile(cdr3VspectratypeCmd.getFile("vj_usage.tsv"), "tsv", vjUsageConv.getColumns())
+	outputs["vjUsagePf"] = pframes.exportFrame(vjUsagePf)
 
 	return {
 		outputs: outputs,

package/src/pf-spectratype-conv.lib.tengo

@@ -0,0 +1,59 @@
+ll := import("@platforma-sdk/workflow-tengo:ll")
+
+getColumns := func() {
+    return {
+        "axes": [
+            {
+                "column": "chain",
+                "spec": {
+                    "name": "pl7.app/vdj/chain",
+                    "type": "String",
+                    "annotations": { "pl7.app/label": "CDR3 chain" }
+                }
+            },
+            {
+                "column": "cdr3Length",
+                "spec": {
+                    "name": "pl7.app/vdj/sequenceLength",
+                    "type": "Int",
+                    "domain": {
+                        "pl7.app/vdj/feature": "CDR3",
+                        "pl7.app/alphabet": "aminoacid"
+                    },
+                    "annotations": { "pl7.app/label": "CDR3 aa Length" }
+                }
+            },
+            {
+                "column": "vGene",
+                "spec": {
+                    "name": "pl7.app/vdj/geneHit",
+                    "type": "String",
+                    "domain": { "pl7.app/vdj/reference": "VGene" },
+                    "annotations": {
+                        "pl7.app/label": "Best V gene"
+                    }
+                }
+            }
+        ],
+        "columns": [
+            {
+                "column": "count",
+                "spec": {
+                    "name": "pl7.app/vdj/vSpectratype",
+                    "valueType": "Int",
+                    "domain": {
+                        "pl7.app/vdj/feature": "CDR3",
+                        "pl7.app/alphabet": "aminoacid"
+                    },
+                    "annotations": { "pl7.app/label": "CDR3 V Spectratype" }
+                }
+            }
+        ],
+        "storageFormat": "Binary",
+        "partitionKeyLength": 0
+    }
+}
+
+export ll.toStrict({
+    getColumns: getColumns
+})

package/src/pf-vj-usage-conv.lib.tengo

@@ -0,0 +1,54 @@
+ll := import("@platforma-sdk/workflow-tengo:ll")
+
+getColumns := func() {
+    return {
+        "axes": [
+                {
+                "column": "chain",
+                "spec": {
+                    "name": "pl7.app/vdj/chain",
+                    "type": "String",
+                    "annotations": { "pl7.app/label": "CDR3 chain" }
+                }
+            },
+            {
+                "column": "vGene",
+                "spec": {
+                    "name": "pl7.app/vdj/geneHit",
+                    "type": "String",
+                    "domain": { "pl7.app/vdj/reference": "VGene" },
+                    "annotations": {
+                        "pl7.app/label": "Best V gene"
+                    }
+                }
+            },
+            {
+                "column": "jGene",
+                "spec": {
+                    "name": "pl7.app/vdj/geneHit",
+                    "type": "String",
+                    "domain": { "pl7.app/vdj/reference": "JGene" },
+                    "annotations": {
+                        "pl7.app/label": "Best J gene"
+                    }
+                }
+            }
+        ],
+        "columns": [
+            {
+                "column": "count",
+                "spec": {
+                    "name": "pl7.app/vdj/vjGeneUsage",
+                    "valueType": "Int",
+                    "annotations": { "pl7.app/label": "V/J usage" }
+                }
+            }
+        ],
+        "storageFormat": "Binary",
+        "partitionKeyLength": 0
+    }
+}
+
+export ll.toStrict({
+    getColumns: getColumns
+})