@platforma-open/milaboratories.top-antibodies.workflow 3.0.1 → 4.0.0

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@3.0.1 build /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow
+> @platforma-open/milaboratories.top-antibodies.workflow@4.0.0 build /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow
 > shx rm -rf dist && pl-tengo check && pl-tengo build
 
 Processing "src/assembling-fasta.tpl.tengo"...

package/CHANGELOG.md

@@ -1,5 +1,11 @@
 # @platforma-open/milaboratories.top-antibodies.workflow
 
+## 4.0.0
+
+### Major Changes
+
+- 1c1c7c1: Support peptides
+
 ## 3.0.1
 
 ### Patch Changes

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

@@ -6,7 +6,7 @@ getColumns := func(datasetSpec, addRanking, addInVivoScore) {
         id: "link",
         allowNA: false,
         spec: {
-          name: "pl7.app/vdj/lead-selection",
+          name: "pl7.app/lead-selection",
           valueType: "Int",
           domain: {},
           annotations: {
@@ -21,7 +21,7 @@ getColumns := func(datasetSpec, addRanking, addInVivoScore) {
     columns = columns + [{
         column: "ranked_order",
         spec: {
-          name: "pl7.app/vdj/ranking-order",
+          name: "pl7.app/ranking-order",
           valueType: "Int",
           domain: {},
           annotations: {

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

@@ -18,7 +18,7 @@ getColumns := func(datasetSpec, filterMap, rankingMap, diversificationColumn, to
         id: "link",
         allowNA: false,
         spec: {
-          name: "pl7.app/vdj/lead-selection",
+          name: "pl7.app/lead-selection",
           valueType: "Int",
           domain: domain,
           annotations: {

package/dist/tengo/lib/utils.lib.tengo

@@ -61,7 +61,7 @@ findMatchingLinkerIndex := func(colsSpec, linkerColumns) {
 
     rankingClusterIdAxis := undefined
     for axis in colsSpec.axesSpec {
-        if axis.name == "pl7.app/vdj/clusterId" {
+        if axis.name == "pl7.app/clusterId" {
             rankingClusterIdAxis = axis
             break
         }
@@ -76,7 +76,7 @@ findMatchingLinkerIndex := func(colsSpec, linkerColumns) {
 
         linkerClusterIdAxis := undefined
         for axis in linkerCol.spec.axesSpec {
-            if axis.name == "pl7.app/vdj/clusterId" {
+            if axis.name == "pl7.app/clusterId" {
                 linkerClusterIdAxis = axis
                 break
             }
@@ -195,7 +195,7 @@ resolveClusterColumnHeader := func(args, columns, sortedLinkers) {
 
     selectedClusterIdAxis := undefined
     for axis in selectedLinkerSpec.axesSpec {
-        if axis.name == "pl7.app/vdj/clusterId" {
+        if axis.name == "pl7.app/clusterId" {
             selectedClusterIdAxis = axis
             break
         }
@@ -209,7 +209,7 @@ resolveClusterColumnHeader := func(args, columns, sortedLinkers) {
     for linkerIdx, col in sortedLinkers {
 
         for axis in col.spec.axesSpec {
-            if axis.name == "pl7.app/vdj/clusterId" {
+            if axis.name == "pl7.app/clusterId" {
 
                 if clusterAxisDomainsMatch(selectedClusterIdAxis, axis) {
                     return "clusterAxis_" + string(linkerIdx) + "_0"
@@ -409,7 +409,7 @@ initializeCloneTable := func(pframes, columns, args, datasetSpec) {
             addedCols = true
         }
 
-        if !is_undefined(clusterIdAxis) && clusterIdAxis.name == "pl7.app/vdj/clusterId" {
+        if !is_undefined(clusterIdAxis) && clusterIdAxis.name == "pl7.app/clusterId" {
             linkerClusterIdAxesWithIdx = append(linkerClusterIdAxesWithIdx, {
                 axis: clusterIdAxis,
                 linkerIdx: linkerIdx
@@ -423,7 +423,7 @@ initializeCloneTable := func(pframes, columns, args, datasetSpec) {
 
             clusterSizeClusterIdAxis := undefined
             for axis in col.spec.axesSpec {
-                if axis.name == "pl7.app/vdj/clusterId" {
+                if axis.name == "pl7.app/clusterId" {
                     clusterSizeClusterIdAxis = axis
                     break
                 }

package/package.json

@@ -1,19 +1,19 @@
 {
   "name": "@platforma-open/milaboratories.top-antibodies.workflow",
-  "version": "3.0.1",
+  "version": "4.0.0",
   "type": "module",
   "description": "Block Workflow",
   "dependencies": {
-    "@platforma-sdk/workflow-tengo": "5.16.0",
+    "@platforma-sdk/workflow-tengo": "5.20.0",
     "@platforma-open/milaboratories.software-anarci": "^0.0.3",
-    "@platforma-open/milaboratories.top-antibodies.spectratype": "1.8.3",
     "@platforma-open/milaboratories.top-antibodies.sample-clonotypes": "2.1.2",
-    "@platforma-open/milaboratories.top-antibodies.umap": "1.2.3",
     "@platforma-open/milaboratories.top-antibodies.assembling-fasta": "1.3.2",
-    "@platforma-open/milaboratories.top-antibodies.anarci-kabat": "1.4.3"
+    "@platforma-open/milaboratories.top-antibodies.anarci-kabat": "1.4.3",
+    "@platforma-open/milaboratories.top-antibodies.spectratype": "1.8.3",
+    "@platforma-open/milaboratories.top-antibodies.umap": "1.2.3"
   },
   "devDependencies": {
-    "@platforma-sdk/tengo-builder": "2.5.17"
+    "@platforma-sdk/tengo-builder": "2.5.20"
   },
   "scripts": {
     "build": "shx rm -rf dist && pl-tengo check && pl-tengo build",

package/src/main.tpl.tengo

@@ -134,6 +134,7 @@ wf.body(func(args) {
 
         // Needed conditional variable
 	    isSingleCell := datasetSpec.axesSpec[1].name == "pl7.app/vdj/scClonotypeKey"
+	    isPeptide := datasetSpec.axesSpec[1].name == "pl7.app/variantKey"
     
         ////////// Clonotype Filtering //////////
         // Initialize and build clone table with all columns
@@ -177,76 +178,79 @@ wf.body(func(args) {
             // Export only the sampling column for downstream (built inside sub-template to avoid hang)
             exports["sampledColumnsPf"] = filterSampleResult.output("sampledColumnsExport", 24 * 60 * 60 * 1000)
 
-            ////////// CDR3 Length Calculation //////////
-            // Initialize and build CDR3 sequence table
-            cdr3SeqTableBuilt := utils.initializeCdr3SeqTable(pframes, columns, datasetSpec, isSingleCell)
+            // CDR3 spectratype, V/J gene usage, and Kabat numbering are VDJ-only — skip for peptide inputs.
+            if !isPeptide {
+                ////////// CDR3 Length Calculation //////////
+                // Initialize and build CDR3 sequence table
+                cdr3SeqTableBuilt := utils.initializeCdr3SeqTable(pframes, columns, datasetSpec, isSingleCell)
 
-            cdr3VspectratypeCmd := exec.builder().
-                software(assets.importSoftware("@platforma-open/milaboratories.top-antibodies.spectratype:main")).
-                mem("16GiB").
-                cpu(1).
-                addFile("cdr3_sequences_input.parquet", cdr3SeqTableBuilt).
-                arg("--input_parquet").arg("cdr3_sequences_input.parquet").
-                arg("--spectratype_tsv").arg("spectratype.tsv").
-                arg("--vj_usage_tsv").arg("vj_usage.tsv") // no dot here
+                cdr3VspectratypeCmd := exec.builder().
+                    software(assets.importSoftware("@platforma-open/milaboratories.top-antibodies.spectratype:main")).
+                    mem("16GiB").
+                    cpu(1).
+                    addFile("cdr3_sequences_input.parquet", cdr3SeqTableBuilt).
+                    arg("--input_parquet").arg("cdr3_sequences_input.parquet").
+                    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 finalClonotypes != undefined {
-                cdr3VspectratypeCmd = cdr3VspectratypeCmd.
-                    arg("--final-clonotypes").arg("finalClonotypes.parquet").
-                    addFile("finalClonotypes.parquet", finalClonotypes)
-            }
+                // Add top clonotypes argument and file to the builder if provided
+                if finalClonotypes != undefined {
+                    cdr3VspectratypeCmd = cdr3VspectratypeCmd.
+                        arg("--final-clonotypes").arg("finalClonotypes.parquet").
+                        addFile("finalClonotypes.parquet", finalClonotypes)
+                }
 
-            cdr3VspectratypeCmd = cdr3VspectratypeCmd. // continue building the command
-                saveFile("spectratype.tsv").
-                saveFile("vj_usage.tsv").
-                printErrStreamToStdout().
-                cache(24 * 60 * 60 * 1000).
-                run()
+                cdr3VspectratypeCmd = cdr3VspectratypeCmd. // continue building the command
+                    saveFile("spectratype.tsv").
+                    saveFile("vj_usage.tsv").
+                    printErrStreamToStdout().
+                    cache(24 * 60 * 60 * 1000).
+                    run()
 
 
-            // Spectratype PFrame structure is [chain][cdr3Length][vGene] -> count
+                // Spectratype PFrame structure is [chain][cdr3Length][vGene] -> count
 
-            cdr3VspectratypePf := xsv.importFile(cdr3VspectratypeCmd.getFile("spectratype.tsv"), 
-                                                "tsv", spectratypeConv.getColumns(),
-                                                {cpu: 1, mem: "16GiB"})
-            outputs["cdr3VspectratypePf"] = pframes.exportFrame(cdr3VspectratypePf) 
+                cdr3VspectratypePf := xsv.importFile(cdr3VspectratypeCmd.getFile("spectratype.tsv"),
+                                                    "tsv", spectratypeConv.getColumns(),
+                                                    {cpu: 1, mem: "16GiB"})
+                outputs["cdr3VspectratypePf"] = pframes.exportFrame(cdr3VspectratypePf)
 
-            // For vjUsage structure is [chain][vGene][jGene] -> count
-            vjUsagePf := xsv.importFile(cdr3VspectratypeCmd.getFile("vj_usage.tsv"), 
-                                        "tsv", vjUsageConv.getColumns(), 
-                                        {cpu: 1, mem: "16GiB"})
-            outputs["vjUsagePf"] = pframes.exportFrame(vjUsagePf)
+                // For vjUsage structure is [chain][vGene][jGene] -> count
+                vjUsagePf := xsv.importFile(cdr3VspectratypeCmd.getFile("vj_usage.tsv"),
+                                            "tsv", vjUsageConv.getColumns(),
+                                            {cpu: 1, mem: "16GiB"})
+                outputs["vjUsagePf"] = pframes.exportFrame(vjUsagePf)
 
-            if args.kabatNumbering == true {
-            ////////// Assembling AA sequences //////////
-                // Initialize and build assembling sequence table
-                assemInit := utils.initializeAssemSeqTable(pframes, columns, datasetSpec, isSingleCell)
-                assemSeqTableBuilt := assemInit.assemSeqTable
-                bulkChain := assemInit.bulkChain
-                seqCols := assemInit.seqCols
+                if args.kabatNumbering == true {
+                    ////////// Assembling AA sequences //////////
+                    // Initialize and build assembling sequence table
+                    assemInit := utils.initializeAssemSeqTable(pframes, columns, datasetSpec, isSingleCell)
+                    assemSeqTableBuilt := assemInit.assemSeqTable
+                    bulkChain := assemInit.bulkChain
+                    seqCols := assemInit.seqCols
 
-                // Convert assembling feature sequences to FASTA via sub-template
-                assemFastaTpl := assets.importTemplate(":assembling-fasta")
-                assem := render.create(assemFastaTpl, {
-                    inputTsv: assemSeqTableBuilt,
-                    keyColumn: "clonotypeKey",
-                    finalClonotypes: finalClonotypes,
-                    isSingleCell: isSingleCell,
-                    bulkChain: bulkChain
-                })
-                //outputs["assemblingAnarci"] = assem.output("anarci", 24 * 60 * 60 * 1000)
-                kabatFile := assem.output("kabat", 24 * 60 * 60 * 1000)
-                // Derive feature name from assembling feature columns (prefer first column's feature)
-                featName := ""
-                if len(seqCols) > 0 {
-                    f := seqCols[0].spec.domain["pl7.app/vdj/feature"]
-                    if f != undefined { featName = f }
+                    // Convert assembling feature sequences to FASTA via sub-template
+                    assemFastaTpl := assets.importTemplate(":assembling-fasta")
+                    assem := render.create(assemFastaTpl, {
+                        inputTsv: assemSeqTableBuilt,
+                        keyColumn: "clonotypeKey",
+                        finalClonotypes: finalClonotypes,
+                        isSingleCell: isSingleCell,
+                        bulkChain: bulkChain
+                    })
+                    //outputs["assemblingAnarci"] = assem.output("anarci", 24 * 60 * 60 * 1000)
+                    kabatFile := assem.output("kabat", 24 * 60 * 60 * 1000)
+                    // Derive feature name from assembling feature columns (prefer first column's feature)
+                    featName := ""
+                    if len(seqCols) > 0 {
+                        f := seqCols[0].spec.domain["pl7.app/vdj/feature"]
+                        if f != undefined { featName = f }
+                    }
+                    // Convert kabat.tsv to PFrame with proper specs (bulk: select heavy/light)
+                    kabatPf := xsv.importFile(kabatFile, "tsv", kabatConv.getColumns(datasetSpec, featName, bulkChain), {cpu: 1, mem: "8GiB"})
+                    outputs["assemblingKabatPf"] = pframes.exportFrame(kabatPf)
+                    outputs["kabatStatsContent"] = assem.output("kabatStats")
                 }
-                // Convert kabat.tsv to PFrame with proper specs (bulk: select heavy/light)
-                kabatPf := xsv.importFile(kabatFile, "tsv", kabatConv.getColumns(datasetSpec, featName, bulkChain), {cpu: 1, mem: "8GiB"})
-                outputs["assemblingKabatPf"] = pframes.exportFrame(kabatPf)
-                outputs["kabatStatsContent"] = assem.output("kabatStats")
             }
         }
     }

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

@@ -6,7 +6,7 @@ getColumns := func(datasetSpec, addRanking, addInVivoScore) {
         id: "link",
         allowNA: false,
         spec: {
-          name: "pl7.app/vdj/lead-selection",
+          name: "pl7.app/lead-selection",
           valueType: "Int",
           domain: {},
           annotations: {
@@ -21,7 +21,7 @@ getColumns := func(datasetSpec, addRanking, addInVivoScore) {
     columns = columns + [{
         column: "ranked_order",
         spec: {
-          name: "pl7.app/vdj/ranking-order",
+          name: "pl7.app/ranking-order",
           valueType: "Int",
           domain: {},
           annotations: {

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

@@ -18,7 +18,7 @@ getColumns := func(datasetSpec, filterMap, rankingMap, diversificationColumn, to
         id: "link",
         allowNA: false,
         spec: {
-          name: "pl7.app/vdj/lead-selection",
+          name: "pl7.app/lead-selection",
           valueType: "Int",
           domain: domain,
           annotations: {

package/src/utils.lib.tengo

@@ -61,7 +61,7 @@ findMatchingLinkerIndex := func(colsSpec, linkerColumns) {
     // Find the clusterId axis in the ranking column
     rankingClusterIdAxis := undefined
     for axis in colsSpec.axesSpec {
-        if axis.name == "pl7.app/vdj/clusterId" {
+        if axis.name == "pl7.app/clusterId" {
             rankingClusterIdAxis = axis
             break
         }
@@ -76,7 +76,7 @@ findMatchingLinkerIndex := func(colsSpec, linkerColumns) {
         // Get the clusterId axis from the linker column
         linkerClusterIdAxis := undefined
         for axis in linkerCol.spec.axesSpec {
-            if axis.name == "pl7.app/vdj/clusterId" {
+            if axis.name == "pl7.app/clusterId" {
                 linkerClusterIdAxis = axis
                 break
             }
@@ -195,7 +195,7 @@ resolveClusterColumnHeader := func(args, columns, sortedLinkers) {
     // Find the clusterId axis in the selected linker
     selectedClusterIdAxis := undefined
     for axis in selectedLinkerSpec.axesSpec {
-        if axis.name == "pl7.app/vdj/clusterId" {
+        if axis.name == "pl7.app/clusterId" {
             selectedClusterIdAxis = axis
             break
         }
@@ -209,7 +209,7 @@ resolveClusterColumnHeader := func(args, columns, sortedLinkers) {
     for linkerIdx, col in sortedLinkers {
         // Get the clusterId axis from this linker
         for axis in col.spec.axesSpec {
-            if axis.name == "pl7.app/vdj/clusterId" {
+            if axis.name == "pl7.app/clusterId" {
                 // Use clusterAxisDomainsMatch for proper domain comparison
                 if clusterAxisDomainsMatch(selectedClusterIdAxis, axis) {
                     return "clusterAxis_" + string(linkerIdx) + "_0"
@@ -409,7 +409,7 @@ initializeCloneTable := func(pframes, columns, args, datasetSpec) {
             addedCols = true
         }
         // Collect clusterId axes from linker columns to match cluster size columns
-        if !is_undefined(clusterIdAxis) && clusterIdAxis.name == "pl7.app/vdj/clusterId" {
+        if !is_undefined(clusterIdAxis) && clusterIdAxis.name == "pl7.app/clusterId" {
             linkerClusterIdAxesWithIdx = append(linkerClusterIdAxesWithIdx, {
                 axis: clusterIdAxis,
                 linkerIdx: linkerIdx
@@ -423,7 +423,7 @@ initializeCloneTable := func(pframes, columns, args, datasetSpec) {
             // Find the clusterId axis in this cluster size column
             clusterSizeClusterIdAxis := undefined
             for axis in col.spec.axesSpec {
-                if axis.name == "pl7.app/vdj/clusterId" {
+                if axis.name == "pl7.app/clusterId" {
                     clusterSizeClusterIdAxis = axis
                     break
                 }