@platforma-open/milaboratories.mixcr-clonotyping-2.workflow 2.17.0 → 2.18.0

package/.turbo/turbo-build.log

@@ -1,6 +1,6 @@
  WARN  Issue while reading "/home/runner/work/mixcr-clonotyping/mixcr-clonotyping/.npmrc". Failed to replace env in config: ${NPMJS_TOKEN}
 
-> @platforma-open/milaboratories.mixcr-clonotyping-2.workflow@2.17.0 build /home/runner/work/mixcr-clonotyping/mixcr-clonotyping/workflow
+> @platforma-open/milaboratories.mixcr-clonotyping-2.workflow@2.18.0 build /home/runner/work/mixcr-clonotyping/mixcr-clonotyping/workflow
 > rm -rf dist && pl-tengo check && pl-tengo build
 
   info: Skipping unknown file type: test/columns.test.ts

package/CHANGELOG.md

@@ -1,5 +1,14 @@
 # @platforma-open/milaboratories.mixcr-clonotyping.workflow
 
+## 2.18.0
+
+### Minor Changes
+
+- 2a9ce2a: - added two aggregated columns: Mean Fraction Of UMIs/Reads, Supprting UMIs/Reads
+  - support for RNA-Seq preset
+  - fallback for in-frame features for amino-acid sequence columns (for FR4 and VDJRegion)
+  - annotation strings for nucleotide and amino-acid sequences of assembling feature
+
 ## 2.17.0
 
 ### Minor Changes

package/dist/tengo/lib/calculate-export-specs.lib.tengo

@@ -7,10 +7,24 @@ json := import("json")
 a := func(order, defaultVisibility, spec) {
 	return maps.merge(spec, {
 		"pl7.app/table/orderPriority": string(order),
-		"pl7.app/table/visibility": defaultVisibility ? "default" : "optional"
+		"pl7.app/table/visibility": is_undefined(defaultVisibility) ? "hidden" : defaultVisibility ? "default" : "optional"
 	})
 }
 
+
+
+
+
+inFrameFeatures := {
+	"FR4": "FR4InFrame",
+	"VDJRegion": "VDJRegionInFrame"
+}
+
+annotationMappings := {
+	"CDRs": "{\"1\":\"CDR1\",\"2\":\"CDR2\",\"3\":\"CDR3\"}",
+	"Segments": "{\"1\":\"V\",\"2\":\"D\",\"3\":\"J\",\"4\":\"C\"}"
+}
+
 toCombinedDomainValue := func(spec) {
 	result := [spec.name]
 
@@ -42,7 +56,7 @@ assemblingFeatureInfo := func(assemblingFeature) {
 		if assemblingFeature == "CDR3" {
 			productiveFeature = "CDR3"
 		} else if assemblingFeature == "VDJRegion" {
-			productiveFeature = "VDJRegion(0,-1)"
+			productiveFeature = "VDJRegionInFrame"
 			coreVFeature = "{FR1Begin:FR3End}"
 			coreJFeature = "FR4"
 		} else if len(splittedFeature1) == 2 {
@@ -116,8 +130,31 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 	productiveFeature := assemblingFeatureInfo.productiveFeature
 	coreGeneFeatures := assemblingFeatureInfo.coreGeneFeatures
 
+
+	anchorFeature := undefined;
+
+	features := undefined
+	if is_undefined(assemblingFeature) {
+		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4"]
+
+
+
+
+
+		assemblingFeature = "CDR3"
+
+		anchorFeature = "CDR3"
+	} else if assemblingFeature != "CDR3" {
+		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4", assemblingFeature]
+		anchorFeature = assemblingFeature
+	} else {
+		features = ["CDR3"]
+		anchorFeature = "CDR3"
+	}
+
 	clonotypeKeyColumns := undefined
 	clonotypeKeyArgs := undefined
+
 	if !is_undefined(assemblingFeature) {
 		clonotypeKeyColumns = ["nSeq" + assemblingFeature, "bestVGene", "bestJGene"]
 		clonotypeKeyArgs = [
@@ -129,7 +166,6 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 			clonotypeKeyColumns += ["bestCGene"]
 			clonotypeKeyArgs += [ [ "-cGene" ] ]
 		}
-
 	}
 
 	columnsSpecPerSample := []
@@ -181,7 +217,42 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 		[ "-readCount" ],
 		[ "-readFraction" ]
 	]
-	mainAbundanceColumn := "readFraction"
+
+	mainAbundanceColumnUnnormalized := "readCount"
+	mainAbundanceColumnNormalized := "readFraction"
+
+	mainAbundanceColumnAggregates := [{
+		column: mainAbundanceColumnUnnormalized + "Sum",
+		id: "read-count-total",
+		allowNA: false,
+		spec: {
+			name: "pl7.app/vdj/readCountTotal",
+			valueType: "Int",
+			annotations: a(87120, true, {
+				"pl7.app/min": "1",
+				"pl7.app/isAbundance": "true",
+				"pl7.app/abundance/unit": "reads",
+				"pl7.app/abundance/normalized": "false",
+				"pl7.app/label": "Supporting Reads"
+			})
+		}
+	}, {
+		column: mainAbundanceColumnNormalized + "Mean",
+		id: "read-fraction-mean",
+		allowNA: false,
+		spec: {
+			name: "pl7.app/vdj/readFractionMean",
+			valueType: "Double",
+			annotations: a(87130, true, {
+				"pl7.app/min": "0",
+				"pl7.app/max": "1",
+				"pl7.app/isAbundance": "true",
+				"pl7.app/abundance/unit": "reads",
+				"pl7.app/abundance/normalized": "true",
+				"pl7.app/label": "Mean Fraction of Reads"
+			})
+		}
+	}]
 
 	if hasUmi {
 		columnsSpecPerSample += [ {
@@ -223,7 +294,40 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 			[ "-uniqueTagCount", "Molecule" ],
 			[ "-uniqueTagFraction", "Molecule" ]
 		]
-		mainAbundanceColumn = "uniqueMoleculeFraction"
+		mainAbundanceColumnNormalized = "uniqueMoleculeFraction"
+		mainAbundanceColumnUnnormalized = "uniqueMoleculeCount"
+		mainAbundanceColumnAggregates = [ {
+			column: mainAbundanceColumnUnnormalized + "Sum",
+			id: "umi-count-total",
+			allowNA: false,
+			spec: {
+				name: "pl7.app/vdj/uniqueMoleculeCountTotal",
+				valueType: "Long",
+				annotations: a(87120, true, {
+					"pl7.app/min": "1",
+					"pl7.app/isAbundance": "true",
+					"pl7.app/abundance/unit": "molecules",
+					"pl7.app/abundance/normalized": "false",
+					"pl7.app/label": "Supporting UMIs"
+				})
+			}
+		}, {
+			column: mainAbundanceColumnNormalized + "Mean",
+			id: "umi-fraction-mean",
+			allowNA: false,
+			spec: {
+				name: "pl7.app/vdj/uniqueMoleculeFractionMean",
+				valueType: "Double",
+				annotations: a(87130, true, {
+					"pl7.app/min": "0",
+					"pl7.app/max": "1",
+					"pl7.app/isAbundance": "true",
+					"pl7.app/abundance/unit": "molecules",
+					"pl7.app/abundance/normalized": "true",
+					"pl7.app/label": "Mean Fraction of UMIs"
+				})
+			}
+		} ]
 	}
 
 	sampleCountColumn := {
@@ -291,6 +395,7 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 		columnsSpecPerClonotypeSc = [ sampleCountColumn ]
 	} else {
 		columnsSpecPerClonotype += [ sampleCountColumn ]
+		columnsSpecPerClonotype += mainAbundanceColumnAggregates
 	}
 
 	orderP := 80000
@@ -298,19 +403,8 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 
 
 
-	anchorFeature := undefined;
 
-	features := undefined
-	if is_undefined(assemblingFeature) {
-		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4"]
-		anchorFeature = "CDR3"
-	} else if assemblingFeature != "CDR3" {
-		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4", assemblingFeature]
-		anchorFeature = assemblingFeature
-	} else {
-		features = ["CDR3"]
-		anchorFeature = "CDR3"
-	}
+	annotationTypes := assemblingFeature == "CDR3" ? ["Segments"] : ["CDRs", "Segments"]
 
 	for isImputed in ( is_undefined(assemblingFeature) ? [false, true] : [false] ) {
 		imputedU := isImputed ? "Imputed" : ""
@@ -318,33 +412,68 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 		for featureU in features {
 			featureL := text.to_lower(formatId(featureU))
 			for isAminoAcid in [true, false] {
+				featureInFrameU := isAminoAcid ? inFrameFeatures[featureU] : featureU
+				if is_undefined(featureInFrameU) {
+					featureInFrameU = featureU
+				}
+				featureInFrameL := text.to_lower(formatId(featureInFrameU))
+
 				alphabet := isAminoAcid ? "aminoacid" : "nucleotide"
 				alphabetShort := isAminoAcid ? "aa" : "nt"
 				alphabetShortMixcr := isAminoAcid ? "aa" : "n"
-				columnName := alphabetShortMixcr + "Seq" + imputedU + featureU
+				columnName := alphabetShortMixcr + "Seq" + imputedU + featureInFrameU
 				visibility := featureU == "CDR3" && (!isSingleCell || isAminoAcid) // isSingleCell ? (featureU == "CDR3") && isAminoAcid : (featureU == "CDR3") || (featureU == assemblingFeature)
 				columnsSpecPerClonotype += [ {
 						column: columnName,
-						id: alphabetShortMixcr + "-seq-" + featureL + (isImputed ? "-imputed" : ""),
+						id: alphabetShortMixcr + "-seq-" + featureInFrameL + (isImputed ? "-imputed" : ""),
 						naRegex: "region_not_covered",
 						spec: {
 							name: "pl7.app/vdj/sequence",
 							valueType: "String",
 							domain: {
-								"pl7.app/vdj/feature": featureU,
+								"pl7.app/vdj/feature": featureInFrameU,
 								"pl7.app/alphabet": alphabet
 							},
 							annotations: a(orderP, visibility, {
 								"pl7.app/vdj/isAssemblingFeature": featureU == anchorFeature ? "true" : "false",
+								"pl7.app/vdj/isMainSequence": featureU == anchorFeature ? "true" : "false",
 								"pl7.app/vdj/imputed": string(isImputed),
-								"pl7.app/label": featureU + " " + alphabetShort
+								"pl7.app/label": featureInFrameU + " " + alphabetShort
 							})
 						}
 					} ]
-				exportArgs += [ [ "-" + alphabetShortMixcr + "Feature" + imputedU, featureU ] ]
+				exportArgs += [ [ "-" + alphabetShortMixcr + "Feature" + imputedU, featureInFrameU ] ]
 				orderP -= 100
 
 
+				if !isImputed && featureU == assemblingFeature {
+					for annotationType in annotationTypes {
+						columnName := alphabetShortMixcr + "AnnotationOf" + annotationType + "For" + featureInFrameU
+						columnsSpecPerClonotype += [ {
+							column: columnName,
+							id: alphabetShortMixcr + "-annotation-" + annotationType + "-" + featureInFrameL,
+							naRegex: "region_not_covered",
+							spec: {
+								name: "pl7.app/vdj/sequence/annotation",
+								valueType: "String",
+								domain: {
+									"pl7.app/vdj/feature": featureInFrameU,
+									"pl7.app/alphabet": alphabet,
+									"pl7.app/sequence/annotation/type": annotationType
+								},
+								annotations: a(orderP, undefined, {
+									"pl7.app/label": annotationType + " annotation for " + featureInFrameU + " " + alphabetShort,
+									"pl7.app/sequence/annotation/mapping": annotationMappings[annotationType],
+									"pl7.app/sequence/isAnnotation": "true"
+								})
+							}
+						} ]
+						exportArgs += [ [ "-" + alphabetShortMixcr + "AnnotationString", annotationType, featureInFrameU ] ]
+						orderP -= 100
+					}
+				}
+
+
 				if featureU == "CDR3" {
 					columnsSpecPerClonotype += [ {
 						column: alphabetShortMixcr + "Length" + featureU,
@@ -701,7 +830,9 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 
 		columnsSpec: columnsSpec,
 
-		mainAbundanceColumn: mainAbundanceColumn,
+		mainAbundanceColumnNormalized: mainAbundanceColumnNormalized,
+		mainAbundanceColumnUnnormalized: mainAbundanceColumnUnnormalized,
+
 		mainProductiveColumn: mainProductiveColumn,
 		mainProductiveArgs: mainProductiveArgs,
 

package/package.json

@@ -1,17 +1,17 @@
 {
   "name": "@platforma-open/milaboratories.mixcr-clonotyping-2.workflow",
-  "version": "2.17.0",
+  "version": "2.18.0",
   "description": "Tengo-based template",
   "dependencies": {
-    "@platforma-sdk/workflow-tengo": "^4.1.2",
+    "@platforma-sdk/workflow-tengo": "^4.3.2",
     "@platforma-open/milaboratories.mixcr-clonotyping-2.single-cell-scripts": "1.1.3",
     "@platforma-open/milaboratories.mixcr-clonotyping-2.hash-column": "1.1.1"
   },
   "devDependencies": {
-    "@platforma-sdk/tengo-builder": "^2.1.0",
-    "@platforma-open/milaboratories.software-mixcr": "4.7.0-165-develop",
+    "@platforma-sdk/tengo-builder": "^2.1.3",
+    "@platforma-open/milaboratories.software-mixcr": "4.7.0-169-develop",
     "@platforma-open/milaboratories.software-ptransform": "^1.4.2",
-    "@platforma-sdk/test": "^1.29.17",
+    "@platforma-sdk/test": "^1.30.24",
     "vitest": "~2.1.8",
     "typescript": "~5.5.4"
   },

package/src/aggregate-by-clonotype-key.tpl.tengo

@@ -19,12 +19,13 @@ self.body(func(inputs) {
 	ll.assert(inputDataMeta.keyLength == 1, "unexpected number of aggregation axes")
 
 	params := inputs.params
-	mainAbundanceColumn := params.mainAbundanceColumn
+	mainAbundanceColumnNormalized := params.mainAbundanceColumnNormalized
+	mainAbundanceColumnUnnormalized := params.mainAbundanceColumnUnnormalized
 	clonotypeColumns := params.clonotypeColumns
 
 	pickCols := []
 	for col in clonotypeColumns {
-		if col == "sampleCount" {
+		if col == "sampleCount" || col == mainAbundanceColumnNormalized + "Mean" || col == mainAbundanceColumnUnnormalized + "Sum" {
 			continue
 		}
 		pickCols = append(pickCols, [col, col])
@@ -37,13 +38,21 @@ self.body(func(inputs) {
 			groupBy: ["clonotypeKey"],
 			aggregations: [ {
 				type: "max_by",
-				rankingCol: mainAbundanceColumn,
+				rankingCol: mainAbundanceColumnNormalized,
 				pickCols: pickCols
 			}, {
 				type: "count",
-				src: mainAbundanceColumn,
+				src: mainAbundanceColumnNormalized,
 				dst: "sampleCount"
-			} ]
+			}, {
+				type: "sum",
+				src: mainAbundanceColumnUnnormalized,
+				dst: mainAbundanceColumnUnnormalized + "Sum"
+			}, {
+				type: "mean",
+				src: mainAbundanceColumnNormalized,
+				dst: mainAbundanceColumnNormalized + "Mean"
+			}]
 		} ]
 	}
 

package/src/calculate-export-specs.lib.tengo

@@ -7,10 +7,24 @@ json := import("json")
 a := func(order, defaultVisibility, spec) {
 	return maps.merge(spec, {
 		"pl7.app/table/orderPriority": string(order),
-		"pl7.app/table/visibility": defaultVisibility ? "default" : "optional"
+		"pl7.app/table/visibility": is_undefined(defaultVisibility) ? "hidden" : defaultVisibility ? "default" : "optional"
 	})
 }
 
+/**
+ * Mapping from original featuers to their in-frame variants when available.
+ * This is used to avoid showing non-informative features in the clonotype browser.
+ */
+inFrameFeatures := {
+	"FR4": "FR4InFrame",
+	"VDJRegion": "VDJRegionInFrame"
+}
+
+annotationMappings := {
+	"CDRs": "{\"1\":\"CDR1\",\"2\":\"CDR2\",\"3\":\"CDR3\"}",
+	"Segments": "{\"1\":\"V\",\"2\":\"D\",\"3\":\"J\",\"4\":\"C\"}"
+}
+
 toCombinedDomainValue := func(spec) {
 	result := [spec.name]
 	// getKeys sort keys
@@ -42,7 +56,7 @@ assemblingFeatureInfo := func(assemblingFeature) {
 		if assemblingFeature == "CDR3" {
 			productiveFeature = "CDR3"
 		} else if assemblingFeature == "VDJRegion" {
-			productiveFeature = "VDJRegion(0,-1)"
+			productiveFeature = "VDJRegionInFrame"
 			coreVFeature = "{FR1Begin:FR3End}"
 			coreJFeature = "FR4"
 		} else if len(splittedFeature1) == 2 {
@@ -116,8 +130,31 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 	productiveFeature := assemblingFeatureInfo.productiveFeature
 	coreGeneFeatures := assemblingFeatureInfo.coreGeneFeatures
 
+	// column with nucleotide sequence of this feature will be marked as anchor
+	anchorFeature := undefined;
+
+	features := undefined
+	if is_undefined(assemblingFeature) {
+		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4"]
+
+		// override assembling feature returned by the mixcr, we assume that if it is undefined,
+		// it means that the assembling feature is CDR3, and assemble contigs is executed with null
+		// subcloning region
+		// TODO return more detailed information from MiXCR
+		assemblingFeature = "CDR3"
+
+		anchorFeature = "CDR3"
+	} else if assemblingFeature != "CDR3" {
+		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4", assemblingFeature]
+		anchorFeature = assemblingFeature
+	} else {
+		features = ["CDR3"]
+		anchorFeature = "CDR3"
+	}
+
 	clonotypeKeyColumns := undefined
 	clonotypeKeyArgs := undefined
+
 	if !is_undefined(assemblingFeature) {
 		clonotypeKeyColumns = ["nSeq" + assemblingFeature, "bestVGene", "bestJGene"]
 		clonotypeKeyArgs = [
@@ -129,7 +166,6 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 			clonotypeKeyColumns += ["bestCGene"]
 			clonotypeKeyArgs += [ [ "-cGene" ] ]
 		}
-
 	}
 
 	columnsSpecPerSample := []
@@ -181,7 +217,42 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 		[ "-readCount" ],
 		[ "-readFraction" ]
 	]
-	mainAbundanceColumn := "readFraction"
+
+	mainAbundanceColumnUnnormalized := "readCount"
+	mainAbundanceColumnNormalized := "readFraction"
+
+	mainAbundanceColumnAggregates := [{
+		column: mainAbundanceColumnUnnormalized + "Sum",
+		id: "read-count-total",
+		allowNA: false,
+		spec: {
+			name: "pl7.app/vdj/readCountTotal",
+			valueType: "Int",
+			annotations: a(87120, true, {
+				"pl7.app/min": "1",
+				"pl7.app/isAbundance": "true",
+				"pl7.app/abundance/unit": "reads",
+				"pl7.app/abundance/normalized": "false",
+				"pl7.app/label": "Supporting Reads"
+			})
+		}
+	}, {
+		column: mainAbundanceColumnNormalized + "Mean",
+		id: "read-fraction-mean",
+		allowNA: false,
+		spec: {
+			name: "pl7.app/vdj/readFractionMean",
+			valueType: "Double",
+			annotations: a(87130, true, {
+				"pl7.app/min": "0",
+				"pl7.app/max": "1",
+				"pl7.app/isAbundance": "true",
+				"pl7.app/abundance/unit": "reads",
+				"pl7.app/abundance/normalized": "true",
+				"pl7.app/label": "Mean Fraction of Reads"
+			})
+		}
+	}]
 
 	if hasUmi {
 		columnsSpecPerSample += [ {
@@ -223,7 +294,40 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 			[ "-uniqueTagCount", "Molecule" ],
 			[ "-uniqueTagFraction", "Molecule" ]
 		]
-		mainAbundanceColumn = "uniqueMoleculeFraction"
+		mainAbundanceColumnNormalized = "uniqueMoleculeFraction"
+		mainAbundanceColumnUnnormalized = "uniqueMoleculeCount"
+		mainAbundanceColumnAggregates = [ {
+			column: mainAbundanceColumnUnnormalized + "Sum",
+			id: "umi-count-total",
+			allowNA: false,
+			spec: {
+				name: "pl7.app/vdj/uniqueMoleculeCountTotal",
+				valueType: "Long",
+				annotations: a(87120, true, {
+					"pl7.app/min": "1",
+					"pl7.app/isAbundance": "true",
+					"pl7.app/abundance/unit": "molecules",
+					"pl7.app/abundance/normalized": "false",
+					"pl7.app/label": "Supporting UMIs"
+				})
+			}
+		}, {
+			column: mainAbundanceColumnNormalized + "Mean",
+			id: "umi-fraction-mean",
+			allowNA: false,
+			spec: {
+				name: "pl7.app/vdj/uniqueMoleculeFractionMean",
+				valueType: "Double",
+				annotations: a(87130, true, {
+					"pl7.app/min": "0",
+					"pl7.app/max": "1",
+					"pl7.app/isAbundance": "true",
+					"pl7.app/abundance/unit": "molecules",
+					"pl7.app/abundance/normalized": "true",
+					"pl7.app/label": "Mean Fraction of UMIs"
+				})
+			}
+		} ]
 	}
 
 	sampleCountColumn := {
@@ -291,26 +395,16 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 		columnsSpecPerClonotypeSc = [ sampleCountColumn ]
 	} else {
 		columnsSpecPerClonotype += [ sampleCountColumn ]
+		columnsSpecPerClonotype += mainAbundanceColumnAggregates
 	}
 
 	orderP := 80000
 
 	// Sequences
 
-	// column with nucleotide sequence of this feature will be marked as anchor
-	anchorFeature := undefined;
-
-	features := undefined
-	if is_undefined(assemblingFeature) {
-		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4"]
-		anchorFeature = "CDR3"
-	} else if assemblingFeature != "CDR3" {
-		features = ["CDR1", "FR1", "FR2", "CDR2", "FR3", "CDR3", "FR4", assemblingFeature]
-		anchorFeature = assemblingFeature
-	} else {
-		features = ["CDR3"]
-		anchorFeature = "CDR3"
-	}
+	// nAnnotationOfCDRsForVDJRegionInFrame
+	// aaAnnotationOfSegmentsForVDJRegionInFrame
+	annotationTypes := assemblingFeature == "CDR3" ? ["Segments"] : ["CDRs", "Segments"]
 
 	for isImputed in ( is_undefined(assemblingFeature) ? [false, true] : [false] ) {
 		imputedU := isImputed ? "Imputed" : ""
@@ -318,32 +412,67 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 		for featureU in features {
 			featureL := text.to_lower(formatId(featureU))
 			for isAminoAcid in [true, false] {
+				featureInFrameU := isAminoAcid ? inFrameFeatures[featureU] : featureU
+				if is_undefined(featureInFrameU) {
+					featureInFrameU = featureU
+				}
+				featureInFrameL := text.to_lower(formatId(featureInFrameU))
+
 				alphabet := isAminoAcid ? "aminoacid" : "nucleotide"
 				alphabetShort := isAminoAcid ? "aa" : "nt"
 				alphabetShortMixcr := isAminoAcid ? "aa" : "n"
-				columnName := alphabetShortMixcr + "Seq" + imputedU + featureU
+				columnName := alphabetShortMixcr + "Seq" + imputedU + featureInFrameU
 				visibility := featureU == "CDR3" && (!isSingleCell || isAminoAcid) // isSingleCell ? (featureU == "CDR3") && isAminoAcid : (featureU == "CDR3") || (featureU == assemblingFeature)
 				columnsSpecPerClonotype += [ {
 						column: columnName,
-						id: alphabetShortMixcr + "-seq-" + featureL + (isImputed ? "-imputed" : ""),
+						id: alphabetShortMixcr + "-seq-" + featureInFrameL + (isImputed ? "-imputed" : ""),
 						naRegex: "region_not_covered",
 						spec: {
 							name: "pl7.app/vdj/sequence",
 							valueType: "String",
 							domain: {
-								"pl7.app/vdj/feature": featureU,
+								"pl7.app/vdj/feature": featureInFrameU,
 								"pl7.app/alphabet": alphabet
 							},
 							annotations: a(orderP, visibility, {
 								"pl7.app/vdj/isAssemblingFeature": featureU == anchorFeature ? "true" : "false",
+								"pl7.app/vdj/isMainSequence": featureU == anchorFeature ? "true" : "false",
 								"pl7.app/vdj/imputed": string(isImputed),
-								"pl7.app/label": featureU + " " + alphabetShort
+								"pl7.app/label": featureInFrameU + " " + alphabetShort
 							})
 						}
 					} ]
-				exportArgs += [ [ "-" + alphabetShortMixcr + "Feature" + imputedU, featureU ] ]
+				exportArgs += [ [ "-" + alphabetShortMixcr + "Feature" + imputedU, featureInFrameU ] ]
 				orderP -= 100
 
+				// Adding sequence annotation columns for assembling feature
+				if !isImputed && featureU == assemblingFeature {
+					for annotationType in annotationTypes {
+						columnName := alphabetShortMixcr + "AnnotationOf" + annotationType + "For" + featureInFrameU
+						columnsSpecPerClonotype += [ {
+							column: columnName,
+							id: alphabetShortMixcr + "-annotation-" + annotationType + "-" + featureInFrameL,
+							naRegex: "region_not_covered",
+							spec: {
+								name: "pl7.app/vdj/sequence/annotation",
+								valueType: "String",
+								domain: {
+									"pl7.app/vdj/feature": featureInFrameU,
+									"pl7.app/alphabet": alphabet,
+									"pl7.app/sequence/annotation/type": annotationType
+								},
+								annotations: a(orderP, undefined, {
+									"pl7.app/label": annotationType + " annotation for " + featureInFrameU + " " + alphabetShort,
+									"pl7.app/sequence/annotation/mapping": annotationMappings[annotationType],
+									"pl7.app/sequence/isAnnotation": "true"
+								})
+							}
+						} ]
+						exportArgs += [ [ "-" + alphabetShortMixcr + "AnnotationString", annotationType, featureInFrameU ] ]
+						orderP -= 100
+					}
+				}
+
 				// For now calculate length only for CDR3 to keep the number of columns manageable
 				if featureU == "CDR3" {
 					columnsSpecPerClonotype += [ {
@@ -701,7 +830,9 @@ calculateExportSpecs := func(presetSpecForBack, blockId) {
 
 		columnsSpec: columnsSpec,
 
-		mainAbundanceColumn: mainAbundanceColumn,
+		mainAbundanceColumnNormalized: mainAbundanceColumnNormalized,
+		mainAbundanceColumnUnnormalized: mainAbundanceColumnUnnormalized,
+
 		mainProductiveColumn: mainProductiveColumn,
 		mainProductiveArgs: mainProductiveArgs,
 

package/src/process.tpl.tengo

@@ -214,7 +214,8 @@ self.body(func(inputs) {
 
 	exportArgs := exportSpecs.exportArgs
 
-	mainAbundanceColumn := exportSpecs.mainAbundanceColumn
+	mainAbundanceColumnNormalized := exportSpecs.mainAbundanceColumnNormalized
+	mainAbundanceColumnUnnormalized := exportSpecs.mainAbundanceColumnUnnormalized
 
 	if is_undefined(axesByClonotypeKey) {
 		ll.panic("Absent clonotype key not supported")
@@ -408,7 +409,8 @@ self.body(func(inputs) {
 				aggregate: ["pl7.app/sampleId"],
 				extra: {
 					params: {
-						mainAbundanceColumn: mainAbundanceColumn,
+						mainAbundanceColumnNormalized: mainAbundanceColumnNormalized,
+						mainAbundanceColumnUnnormalized: mainAbundanceColumnUnnormalized,
 						clonotypeColumns: slices.map(columnsSpecPerClonotype, func(col) {
 							return col.column
 						})

package/src/test/columns.test.ts

@@ -59,7 +59,7 @@ const testCases: TestCase[] = [
       expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqCDR2')).toBeDefined();
       expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqFR3')).toBeDefined();
       expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqCDR3')).toBeDefined();
-      expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqFR4')).toBeDefined();
+      expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqFR4InFrame')).toBeDefined();
       expect(config.columnsSpec.find((c: any) => c.column === 'topChains')).toBeDefined();
     }
   },
@@ -96,7 +96,7 @@ const testCases: TestCase[] = [
       expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqCDR2')).toBeDefined();
       expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqFR3')).toBeDefined();
       expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqCDR3')).toBeDefined();
-      expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqFR4')).toBeDefined();
+      expect(config.columnsSpec.find((c: any) => c.column === 'aaSeqFR4InFrame')).toBeDefined();
       expect(config.columnsSpec.find((c: any) => c.column === 'isotypePrimary')).toBeDefined();
     }
   },

package/.turbo/turbo-test.log

@@ -1,21 +0,0 @@
- WARN  Issue while reading "/home/runner/work/mixcr-clonotyping/mixcr-clonotyping/.npmrc". Failed to replace env in config: ${NPMJS_TOKEN}
-
-> @platforma-open/milaboratories.mixcr-clonotyping-2.workflow@2.17.0 test /home/runner/work/mixcr-clonotyping/mixcr-clonotyping/workflow
-> vitest
-
-
- RUN  v2.1.8 /home/runner/work/mixcr-clonotyping/mixcr-clonotyping/workflow
-
- ✓ src/test/columns.test.ts (6 tests) 52852ms
-   ✓ checking preset for 'milab-human-dna-xcr-7genes-multiplex' 13107ms
-   ✓ checking preset for '10x-sc-xcr-vdj' 8442ms
-   ✓ checking preset for 'cellecta-human-rna-xcr-umi-drivermap-…' 7979ms
-   ✓ checking preset for 'takara-human-rna-bcr-umi-smartseq' 7987ms
-   ✓ checking preset for 'rna-seq' 7550ms
-   ✓ checking preset for 'generic-single-cell-gex' 7787ms
-
- Test Files  1 passed (1)
-      Tests  6 passed (6)
-   Start at  15:43:08
-   Duration  53.65s (transform 43ms, setup 0ms, collect 567ms, tests 52.85s, environment 0ms, prepare 58ms)
-