@platforma-open/milaboratories.top-antibodies.workflow 1.11.3 → 1.13.0

package/.turbo/turbo-build.log

@@ -1,19 +1,23 @@
  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@1.11.3 build /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow
+> @platforma-open/milaboratories.top-antibodies.workflow@1.13.0 build /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow
 > rm -rf dist && pl-tengo check && pl-tengo build
 
+Processing "src/assembling-fasta.tpl.tengo"...
 Processing "src/filter-and-sample.tpl.tengo"...
 Processing "src/main.tpl.tengo"...
+Processing "src/pf-kabat-conv.lib.tengo"...
 Processing "src/pf-spectratype-conv.lib.tengo"...
 Processing "src/pf-vj-usage-conv.lib.tengo"...
 Processing "src/prerun.tpl.tengo"...
 Processing "src/sampled-cols-conv.lib.tengo"...
 No syntax errors found.
   info: Compiling 'dist'...
+  info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/lib/pf-kabat-conv.lib.tengo
   info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/lib/pf-spectratype-conv.lib.tengo
   info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/lib/pf-vj-usage-conv.lib.tengo
   info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/lib/sampled-cols-conv.lib.tengo
+  info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/tpl/assembling-fasta.plj.gz
   info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/tpl/filter-and-sample.plj.gz
   info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/tpl/prerun.plj.gz
   info:   - writing /home/runner/work/antibody-tcr-lead-selection/antibody-tcr-lead-selection/workflow/dist/tengo/tpl/main.plj.gz

package/CHANGELOG.md

@@ -1,5 +1,31 @@
 # @platforma-open/milaboratories.top-antibodies.workflow
 
+## 1.13.0
+
+### Minor Changes
+
+- 3825a42: Fix errors related to numeric properties that apply only to a subset of clonotypes and to multiple clustering blocks upstream
+
+### Patch Changes
+
+- Updated dependencies [3825a42]
+  - @platforma-open/milaboratories.top-antibodies.sample-clonotypes@1.6.0
+  - @platforma-open/milaboratories.top-antibodies.spectratype@1.6.0
+
+## 1.12.0
+
+### Minor Changes
+
+- ccc8076: kabat numbering added
+
+### Patch Changes
+
+- Updated dependencies [ccc8076]
+  - @platforma-open/milaboratories.top-antibodies.sample-clonotypes@1.5.0
+  - @platforma-open/milaboratories.top-antibodies.assembling-fasta@1.1.0
+  - @platforma-open/milaboratories.top-antibodies.anarci-kabat@1.1.0
+  - @platforma-open/milaboratories.top-antibodies.spectratype@1.5.0
+
 ## 1.11.3
 
 ### Patch Changes

package/dist/index.cjs

@@ -1,4 +1,5 @@
 module.exports = { Templates: {
+  'assembling-fasta': { type: 'from-file', path: require.resolve('./tengo/tpl/assembling-fasta.plj.gz') },
   'filter-and-sample': { type: 'from-file', path: require.resolve('./tengo/tpl/filter-and-sample.plj.gz') },
   'prerun': { type: 'from-file', path: require.resolve('./tengo/tpl/prerun.plj.gz') },
   'main': { type: 'from-file', path: require.resolve('./tengo/tpl/main.plj.gz') }

package/dist/index.d.ts

@@ -1,4 +1,4 @@
 declare type TemplateFromFile = { readonly type: "from-file"; readonly path: string; };
-declare type TplName = "filter-and-sample" | "prerun" | "main";
+declare type TplName = "assembling-fasta" | "filter-and-sample" | "prerun" | "main";
 declare const Templates: Record<TplName, TemplateFromFile>;
 export { Templates };

package/dist/index.js

@@ -1,5 +1,6 @@
 import { resolve } from 'node:path';
 export const Templates = {
+  'assembling-fasta': { type: 'from-file', path: resolve(import.meta.dirname, './tengo/tpl/assembling-fasta.plj.gz') },
   'filter-and-sample': { type: 'from-file', path: resolve(import.meta.dirname, './tengo/tpl/filter-and-sample.plj.gz') },
   'prerun': { type: 'from-file', path: resolve(import.meta.dirname, './tengo/tpl/prerun.plj.gz') },
   'main': { type: 'from-file', path: resolve(import.meta.dirname, './tengo/tpl/main.plj.gz') }

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

@@ -0,0 +1,131 @@
+ll := import("@platforma-sdk/workflow-tengo:ll")
+
+getColumns := func(datasetSpec, featureName, bulkChain) {
+  isSingleCell := datasetSpec.axesSpec[1].name == "pl7.app/vdj/scClonotypeKey"
+
+
+  featSuf := ""
+  if !is_undefined(featureName) && featureName != "" { featSuf = "." + string(featureName) }
+
+  cols := [
+    {
+      column: "kabatSequence_H",
+      spec: {
+        name: "pl7.app/vdj/kabatSequence" + featSuf,
+        valueType: "String",
+        domain: {
+          "pl7.app/vdj/chain": "IGHeavy"
+        },
+        annotations: {
+          "pl7.app/label": "KABAT sequence " + featureName + " Heavy",
+          "pl7.app/table/orderPriority": "10",
+		      "pl7.app/table/visibility": "default"
+        }
+      }
+    },
+    {
+      column: "kabatPositions_H",
+      spec: {
+        name: "pl7.app/vdj/kabatPositions" + featSuf,
+        valueType: "String",
+        domain: {
+          "pl7.app/vdj/chain": "IGHeavy"
+        },
+        annotations: {
+          "pl7.app/label": "KABAT positions " + featureName + " Heavy",
+          "pl7.app/table/orderPriority": "9",
+		      "pl7.app/table/visibility": "optional"
+        }
+      }
+    }
+  ]
+
+  if isSingleCell {
+    cols = cols + [
+      {
+        column: "kabatSequence_KL",
+        spec: {
+          name: "pl7.app/vdj/kabatSequence" + featSuf,
+          valueType: "String",
+          domain: {
+            "pl7.app/vdj/chain": "IGLight"
+          },
+          annotations: {
+            "pl7.app/label": "KABAT sequence " + featureName + " Light",
+            "pl7.app/table/orderPriority": "8",
+		        "pl7.app/table/visibility": "default"
+          }
+        }
+      },
+      {
+        column: "kabatPositions_KL",
+        spec: {
+          name: "pl7.app/vdj/kabatPositions" + featSuf,
+          valueType: "String",
+          domain: {
+            "pl7.app/vdj/chain": "IGLight"
+          },
+          annotations: {
+            "pl7.app/label": "KABAT positions " + featureName + " Light",
+            "pl7.app/table/orderPriority": "7",
+		        "pl7.app/table/visibility": "optional"
+          }
+        }
+      }
+    ]
+  } else {
+
+    if bulkChain == "KL" {
+      cols = [
+        {
+          column: "kabatSequence_KL",
+          spec: {
+            name: "pl7.app/vdj/kabatSequence" + featSuf,
+            valueType: "String",
+            domain: {
+              "pl7.app/vdj/chain": "IGLight"
+            },
+            annotations: {
+              "pl7.app/label": "KABAT sequence " + featureName + " Light",
+              "pl7.app/table/orderPriority": "8",
+ 		          "pl7.app/table/visibility": "default"
+            }
+          }
+        },
+        {
+          column: "kabatPositions_KL",
+          spec: {
+            name: "pl7.app/vdj/kabatPositions" + featSuf,
+            valueType: "String",
+            domain: {
+              "pl7.app/vdj/chain": "IGLight"
+            },
+            annotations: {
+              "pl7.app/label": "KABAT positions " + featureName + " Light",
+              "pl7.app/table/orderPriority": "7",
+ 		          "pl7.app/table/visibility": "optional"
+            }
+          }
+        }
+      ]
+    }
+  }
+
+  return {
+    axes: [
+      {
+        column: "clonotypeKey",
+        spec: datasetSpec.axesSpec[1]
+      }
+    ],
+    columns: cols,
+    storageFormat: "Parquet",
+    partitionKeyLength: 0
+  }
+}
+
+export ll.toStrict({
+  getColumns: getColumns
+})
+
+

package/package.json

@@ -1,17 +1,20 @@
 {
   "name": "@platforma-open/milaboratories.top-antibodies.workflow",
-  "version": "1.11.3",
+  "version": "1.13.0",
   "type": "module",
   "description": "Block Workflow",
   "dependencies": {
-    "@platforma-sdk/workflow-tengo": "^5.4.2",
-    "@platforma-open/milaboratories.top-antibodies.sample-clonotypes": "1.4.4",
-    "@platforma-open/milaboratories.top-antibodies.spectratype": "1.4.4",
-    "@platforma-open/milaboratories.top-antibodies.umap": "1.1.4"
+    "@platforma-sdk/workflow-tengo": "^5.5.5",
+    "@platforma-open/milaboratories.software-anarci": "^0.0.3",
+    "@platforma-open/milaboratories.top-antibodies.spectratype": "1.6.0",
+    "@platforma-open/milaboratories.top-antibodies.umap": "1.1.4",
+    "@platforma-open/milaboratories.top-antibodies.assembling-fasta": "1.1.0",
+    "@platforma-open/milaboratories.top-antibodies.sample-clonotypes": "1.6.0",
+    "@platforma-open/milaboratories.top-antibodies.anarci-kabat": "1.1.0"
   },
   "devDependencies": {
-    "@platforma-sdk/tengo-builder": "^2.3.2",
-    "@platforma-sdk/test": "^1.44.19",
+    "@platforma-sdk/tengo-builder": "^2.3.3",
+    "@platforma-sdk/test": "^1.45.6",
     "vitest": "^2.1.8"
   },
   "scripts": {

package/src/assembling-fasta.tpl.tengo

@@ -0,0 +1,92 @@
+self := import("@platforma-sdk/workflow-tengo:tpl")
+exec := import("@platforma-sdk/workflow-tengo:exec")
+assets := import("@platforma-sdk/workflow-tengo:assets")
+wf := import("@platforma-sdk/workflow-tengo:workflow")
+pt := import("@platforma-sdk/workflow-tengo:pt")
+pframes := import("@platforma-sdk/workflow-tengo:pframes")
+anarciSw := assets.importSoftware("@platforma-open/milaboratories.software-anarci:main")
+
+self.defineOutputs("kabat")
+
+self.body(func(inputs) {
+
+    inputTsv := inputs.inputTsv
+    keyColumn := inputs.keyColumn // "clonotypeKey" or "scClonotypeKey"
+    finalClonotypesCsv := inputs.finalClonotypesCsv // optional
+    isSingleCell := inputs.isSingleCell // boolean
+    bulkChain := inputs.bulkChain // "H" or "KL" when !isSingleCell
+
+    cmd := exec.builder().
+        software(assets.importSoftware("@platforma-open/milaboratories.top-antibodies.assembling-fasta:main")).
+        cpu(1).
+        mem("4GiB").
+        addFile("assembling.tsv", inputTsv).
+        arg("--input_tsv").arg("assembling.tsv").
+        arg("--key_column").arg(keyColumn).
+        arg("--output_fasta").arg("assembling.fasta")
+
+    if finalClonotypesCsv != undefined {
+        cmd = cmd.addFile("finalClonotypes.csv", finalClonotypesCsv).
+            arg("--final_clonotypes_csv").arg("finalClonotypes.csv")
+    }
+
+    cmd = cmd.saveFile("assembling.fasta").
+        printErrStreamToStdout().
+        saveStdoutContent().
+        cache(24 * 60 * 60 * 1000).
+        run()
+    
+    anarciFileNameBulk := "anarci.csv_" + bulkChain + ".csv"
+    
+     anarciBuilder := exec.builder().
+        software(anarciSw).
+        arg("-i").arg("assembling.fasta").
+        arg("--scheme").arg("kabat").
+        arg("--ncpu").argWithVar("{system.cpu}").
+        arg("-o").arg("anarci.csv").arg("--csv").
+        addFile("assembling.fasta", cmd.getFile("assembling.fasta"))
+    if isSingleCell {
+        anarciBuilder = anarciBuilder.saveFile("anarci.csv_H.csv").saveFile("anarci.csv_KL.csv")
+    } else {
+        anarciBuilder = anarciBuilder.saveFile(anarciFileNameBulk)
+    }
+    anarciBuilder = anarciBuilder.
+        printErrStreamToStdout().
+        saveStdoutContent().
+        cache(24 * 60 * 60 * 1000).
+        run()
+
+    kabatSw := assets.importSoftware("@platforma-open/milaboratories.top-antibodies.anarci-kabat:main")
+    kabatExec := exec.builder().
+        software(kabatSw)
+    if isSingleCell {
+        kabatExec = kabatExec.addFile("anarci.csv_KL.csv", anarciBuilder.getFile("anarci.csv_KL.csv")).
+            arg("--kl_csv").arg("anarci.csv_KL.csv").
+        addFile("anarci.csv_H.csv", anarciBuilder.getFile("anarci.csv_H.csv")).
+            arg("--h_csv").arg("anarci.csv_H.csv")
+    } else {
+        if bulkChain == "H" {
+            kabatExec = kabatExec.addFile("anarci.csv_H.csv", anarciBuilder.getFile("anarci.csv_H.csv")).
+                arg("--h_csv").arg("anarci.csv_H.csv")
+        } else {
+            kabatExec = kabatExec.addFile("anarci.csv_KL.csv", anarciBuilder.getFile("anarci.csv_KL.csv")).
+                arg("--kl_csv").arg("anarci.csv_KL.csv")
+        }
+    }
+    kabatExec = kabatExec.
+        arg("--out_tsv").arg("kabat.tsv").
+        saveFile("kabat.tsv").
+        printErrStreamToStdout().
+        saveStdoutContent().
+        cache(24 * 60 * 60 * 1000).
+        run()
+    
+    kabat := kabatExec.getFile("kabat.tsv")
+
+    return {
+        kabat: kabat
+      //  kabatPf: pframes.exportFrame(kabatDf)
+    }
+})
+
+

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

@@ -0,0 +1,131 @@
+ll := import("@platforma-sdk/workflow-tengo:ll")
+
+getColumns := func(datasetSpec, featureName, bulkChain) {
+  isSingleCell := datasetSpec.axesSpec[1].name == "pl7.app/vdj/scClonotypeKey"
+
+  // Compose feature suffix for spec names
+  featSuf := ""
+  if !is_undefined(featureName) && featureName != "" { featSuf = "." + string(featureName) }
+
+  cols := [
+    {
+      column: "kabatSequence_H",
+      spec: {
+        name: "pl7.app/vdj/kabatSequence" + featSuf,
+        valueType: "String",
+        domain: {
+          "pl7.app/vdj/chain": "IGHeavy"
+        },
+        annotations: {
+          "pl7.app/label": "KABAT sequence " + featureName + " Heavy",
+          "pl7.app/table/orderPriority": "10",
+		      "pl7.app/table/visibility": "default"
+        }
+      }
+    },
+    {
+      column: "kabatPositions_H",
+      spec: {
+        name: "pl7.app/vdj/kabatPositions" + featSuf,
+        valueType: "String",
+        domain: {
+          "pl7.app/vdj/chain": "IGHeavy"
+        },
+        annotations: {
+          "pl7.app/label": "KABAT positions " + featureName + " Heavy",
+          "pl7.app/table/orderPriority": "9",
+		      "pl7.app/table/visibility": "optional"
+        }
+      }
+    }
+  ]
+
+  if isSingleCell {
+    cols = cols + [
+      {
+        column: "kabatSequence_KL",
+        spec: {
+          name: "pl7.app/vdj/kabatSequence" + featSuf,
+          valueType: "String",
+          domain: {
+            "pl7.app/vdj/chain": "IGLight"
+          },
+          annotations: {
+            "pl7.app/label": "KABAT sequence " + featureName + " Light",
+            "pl7.app/table/orderPriority": "8",
+		        "pl7.app/table/visibility": "default"
+          }
+        }
+      },
+      {
+        column: "kabatPositions_KL",
+        spec: {
+          name: "pl7.app/vdj/kabatPositions" + featSuf,
+          valueType: "String",
+          domain: {
+            "pl7.app/vdj/chain": "IGLight"
+          },
+          annotations: {
+            "pl7.app/label": "KABAT positions " + featureName + " Light",
+            "pl7.app/table/orderPriority": "7",
+		        "pl7.app/table/visibility": "optional"
+          }
+        }
+      }
+    ]
+  } else {
+    // bulk: include only heavy or only light according to bulkChain
+    if bulkChain == "KL" {
+      cols = [
+        {
+          column: "kabatSequence_KL",
+          spec: {
+            name: "pl7.app/vdj/kabatSequence" + featSuf,
+            valueType: "String",
+            domain: {
+              "pl7.app/vdj/chain": "IGLight"
+            },
+            annotations: {
+              "pl7.app/label": "KABAT sequence " + featureName + " Light",
+              "pl7.app/table/orderPriority": "8",
+ 		          "pl7.app/table/visibility": "default"
+            }
+          }
+        },
+        {
+          column: "kabatPositions_KL",
+          spec: {
+            name: "pl7.app/vdj/kabatPositions" + featSuf,
+            valueType: "String",
+            domain: {
+              "pl7.app/vdj/chain": "IGLight"
+            },
+            annotations: {
+              "pl7.app/label": "KABAT positions " + featureName + " Light",
+              "pl7.app/table/orderPriority": "7",
+ 		          "pl7.app/table/visibility": "optional"
+            }
+          }
+        }
+      ]
+    }
+  }
+
+  return {
+    axes: [
+      {
+        column: "clonotypeKey",
+        spec: datasetSpec.axesSpec[1]
+      }
+    ],
+    columns: cols,
+    storageFormat: "Parquet",
+    partitionKeyLength: 0
+  }
+}
+
+export ll.toStrict({
+  getColumns: getColumns
+})
+
+

package/src/prerun.tpl.tengo

@@ -7,6 +7,7 @@ pframes := import("@platforma-sdk/workflow-tengo:pframes")
 slices := import("@platforma-sdk/workflow-tengo:slices")
 render := import("@platforma-sdk/workflow-tengo:render")
 ll := import("@platforma-sdk/workflow-tengo:ll")
+kabatConv := import(":pf-kabat-conv")
 
 spectratypeConv := import(":pf-spectratype-conv")
 vjUsageConv := import(":pf-vj-usage-conv")
@@ -97,6 +98,13 @@ wf.prepare(func(args){
 			"pl7.app/vdj/reference": "JGene"
 		}
 	}, "JGenes")
+
+	// Add assembling feature aminoacid sequences (bulk, sc, scFv)
+	bundleBuilder.addMulti({
+		axes: [{ anchor: "main", idx: 1 }], // Clonotype axis
+		annotations: { "pl7.app/vdj/isAssemblingFeature": "true" },
+		domain: { "pl7.app/alphabet": "aminoacid" }
+	}, "assemblingAaSeqs")
     
     return {
         columns: bundleBuilder.build()
@@ -119,7 +127,7 @@ wf.body(func(args) {
         ////////// Clonotype Filtering //////////
         // Build clonotype table
         cloneTable := pframes.csvFileBuilder() 
-        cloneTable.setAxisHeader(datasetSpec.axesSpec[1].name, "clonotypeKey")
+        cloneTable.setAxisHeader(datasetSpec.axesSpec[1], "clonotypeKey")
 
         // Add Filters to table
         addedAxes := []
@@ -142,7 +150,7 @@ wf.body(func(args) {
                     if !slices.hasElement(axesNames, datasetSpec.axesSpec[1].name) {
                         for na, ax in colsSpec.axesSpec {
                             if ax.name != datasetSpec.axesSpec[1].name {
-                                cloneTable.setAxisHeader(ax.name, "cluster_" + string(i) + string(na))
+                                cloneTable.setAxisHeader(ax, "cluster_" + string(i) + string(na))
                                 addedAxes = append(addedAxes, ax.name)
                             }
                         }
@@ -168,7 +176,7 @@ wf.body(func(args) {
                     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) {
-                                cloneTable.setAxisHeader(ax.name, "cluster_" + string(i) + string(na))
+                                cloneTable.setAxisHeader(ax, "cluster_" + string(i) + string(na))
                             }
                         }
                     }	
@@ -192,7 +200,7 @@ wf.body(func(args) {
                 if !slices.hasElement(axesNames, datasetSpec.axesSpec[1].name) {
                     for na, ax in colsSpec.axesSpec {
                         if ax.name != datasetSpec.axesSpec[1].name {
-                            cloneTable.setAxisHeader(ax.name, "cluster_" + string(i) + string(na))
+                            cloneTable.setAxisHeader(ax, "cluster_" + string(i) + string(na))
                         }
                     }
                 }	
@@ -205,11 +213,11 @@ wf.body(func(args) {
             for i, col in columns.getColumns("linkers") {
                 if datasetSpec.axesSpec[1].name == col.spec.axesSpec[1].name {
                     cloneTable.add(col, {header: "linker." + string(i)})
-                    cloneTable.setAxisHeader(col.spec.axesSpec[0].name, "cluster_" + string(i))
+                    cloneTable.setAxisHeader(col.spec.axesSpec[0], "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, {header: "linker." + string(i)})
-                    cloneTable.setAxisHeader(col.spec.axesSpec[1].name, "cluster_" + string(i))
+                    cloneTable.setAxisHeader(col.spec.axesSpec[1], "cluster_" + string(i))
                     linkerAxisSpec["cluster_" + string(i)] = col.spec.axesSpec[1]
                 }
                 addedCols = true
@@ -224,7 +232,7 @@ wf.body(func(args) {
                 // 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.setAxisHeader(axis, "clusterAxis_" + string(i) + "_" + string(axisIdx))
                     }
                 }
             }
@@ -259,7 +267,7 @@ wf.body(func(args) {
             // outputs["sampledRows"] = filterSampleResult.output("sampledRows", 24 * 60 * 60 * 1000)
             
             ////////// CDR3 Length Calculation //////////
-            
+                
             cdr3SeqTable := pframes.tsvFileBuilder()
             cdr3SeqTable.setAxisHeader(datasetSpec.axesSpec[1].name, "clonotypeKey")
 
@@ -295,7 +303,13 @@ wf.body(func(args) {
 
             for col in cdr3Sequences {
                 headerName := makeHeaderName(col, "cdr3Sequence", isSingleCell)
-                cdr3SeqTable.add(col, {header: headerName})
+                if isSingleCell {
+                    if col.spec.domain["pl7.app/vdj/scClonotypeChain/index"] == "primary" {
+                        cdr3SeqTable.add(col, {header: headerName})
+                    }
+                } else {
+                    cdr3SeqTable.add(col, {header: headerName})
+                }
             }
 
             // Process V genes
@@ -355,6 +369,55 @@ wf.body(func(args) {
                                         "tsv", vjUsageConv.getColumns(), 
                                         {cpu: 1, mem: "16GiB"})
             outputs["vjUsagePf"] = pframes.exportFrame(vjUsagePf)
+
+            if args.kabatNumbering == true {
+            ////////// Assembling AA sequences //////////
+                assemSeqTable := pframes.tsvFileBuilder()
+                keyHeader := "clonotypeKey"
+                assemSeqTable.setAxisHeader(datasetSpec.axesSpec[1].name, keyHeader)
+
+                seqCols := columns.getColumns("assemblingAaSeqs")
+                for col in seqCols {
+                    headerName := makeHeaderName(col, "assemblingFeature", isSingleCell)
+                    assemSeqTable.add(col, {header: headerName})
+                }
+
+                assemSeqTable.mem("16GiB")
+                assemSeqTable.cpu(1)
+                assemSeqTableBuilt := assemSeqTable.build()
+
+                // Convert assembling feature sequences to FASTA via sub-template
+                assemFastaTpl := assets.importTemplate(":assembling-fasta")
+                bulkChain := undefined
+                if !isSingleCell {
+                    // infer bulk chain by header names of incoming seq columns (domain uses IGHeavy / IGLight)
+                    chainDetected := "KL"
+                    for col in seqCols {
+                        ch := col.spec.axesSpec[0].domain["pl7.app/vdj/chain"] // e.g., IGHeavy, IGLight
+                        if ch == "IGHeavy" { chainDetected = "H"; break }
+                        if ch == "IGLight" { chainDetected = "KL" }
+                    }
+                    bulkChain = chainDetected
+                }
+                assem := render.create(assemFastaTpl, {
+                    inputTsv: assemSeqTableBuilt,
+                    keyColumn: "clonotypeKey",
+                    finalClonotypesCsv: finalClonotypesCsv,
+                    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)
+            }
         }
     }