@platforma-open/milaboratories.sequence-properties.workflow 1.1.2 → 1.2.1

package/src/columns.lib.tengo

@@ -0,0 +1,446 @@
+// Output PColumn specs emitted by sequence-properties.
+//
+// Two getters wrap a canonical column list and apply per-consumer annotations:
+//
+//   - `forPropertiesPf(args)`     — every column, `pl7.app/isOutput: "true"`
+//                                   added on top of the canonical annotations.
+//   - `forExport(args, blockId)`  — filtered to `pl7.app/isScore: "true"`
+//                                   columns; domain cloned + `pl7.app/blockId`
+//                                   stamped on so downstream blocks can
+//                                   distinguish runs.
+//
+// `args` shape:
+//   { mode, receptor, chains, fullChains, hasFv }
+// — identical to the relevant subset of `process.tpl.tengo`'s `params`.
+
+ll := import("@platforma-sdk/workflow-tengo:ll")
+
+// ΔCharge endpoints — fixed in v1 per spec; the schema accepts additional
+// pH pairs without breaking existing column identities.
+CHARGE_SHIFT_PH_FROM := "7.4"
+CHARGE_SHIFT_PH_TO := "6.0"
+
+// Spec R6b — same description on every chargeShift column (peptide / CDR3 / Fv).
+// Trimmed to ~30 words for the column-header tooltip; PlAgDataTableV2 clips the
+// top edge against the page header. Magnitude rule and scope-exclusion caveats
+// live in pcolumn-spec.md / the spec, not in the tooltip.
+CHARGE_SHIFT_DESC := "Net charge change from pH 7.4 (blood) to pH 6.0 (endosome). Negative values mean the molecule gains positive charge on acidification — the productive direction for histidine-driven pH switching. Histidine dominates (~−0.46 per His)."
+
+// Receptor + chain → human label fragments (CDR3 / full-chain).
+// Spec R13a: PColumn name and chain domain are unchanged; only the label varies.
+labelFragments := func(receptor, chain) {
+	if receptor == "TCRAB" {
+		if chain == "A" { return { cdr3: "CDR-α3", fullChain: "Vα" } }
+		if chain == "B" { return { cdr3: "CDR-β3", fullChain: "Vβ" } }
+	}
+	if receptor == "TCRGD" {
+		if chain == "A" { return { cdr3: "CDR-γ3", fullChain: "Vγ" } }
+		if chain == "B" { return { cdr3: "CDR-δ3", fullChain: "Vδ" } }
+	}
+	// IG / unknown — antibody convention.
+	if chain == "A" { return { cdr3: "CDR-H3", fullChain: "VH" } }
+	return { cdr3: "CDR-L3", fullChain: "VL" }
+}
+
+// Build a single output column descriptor consumed by xsv.importFile.
+// `tsvCol` is the TSV column header emitted by Python (e.g. "charge_peptide", "charge_A_CDR3").
+// Clones the caller's `annotations` dict so the label stamp does not aliasing-
+// leak into any literal a caller reused.
+makeCol := func(tsvCol, valName, valueType, label, domain, annotations) {
+	newAnnotations := {}
+	if annotations {
+		for k, v in annotations {
+			newAnnotations[k] = v
+		}
+	}
+	newAnnotations["pl7.app/label"] = label
+	spec := {
+		name: valName,
+		valueType: valueType,
+		domain: domain,
+		annotations: newAnnotations
+	}
+	return { column: tsvCol, id: tsvCol, naRegex: "", allowNA: true, spec: spec }
+}
+
+// ---------------------------------------------------------------------------
+// Canonical column lists per scope. No `toPlot`, no `blockId` — the public
+// getters layer those on per consumer.
+// ---------------------------------------------------------------------------
+
+buildPeptideColumns := func() {
+	dom := { "pl7.app/feature": "peptide" }
+	cols := []
+
+	cols += [makeCol("charge_peptide", "pl7.app/charge", "Double",
+		"Net Charge (pH 7)", dom, {
+			"pl7.app/format": ".2f",
+			"pl7.app/isScore": "true",
+			"pl7.app/description": "Net charge at pH 7 (Henderson-Hasselbalch, IPC 2.0 peptide pKa set). Positive = net basic (Arg, Lys, His dominate); negative = net acidic (Asp, Glu dominate). No universal preferred direction.",
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "70000"
+		})]
+
+	cols += [makeCol("chargeShift_peptide", "pl7.app/chargeShift", "Double",
+		"Peptide ΔCharge (pH 7.4 → 6.0)", {
+			"pl7.app/feature": "peptide",
+			"pl7.app/pH/from": CHARGE_SHIFT_PH_FROM,
+			"pl7.app/pH/to": CHARGE_SHIFT_PH_TO
+		}, {
+			"pl7.app/format": ".2f",
+			"pl7.app/description": CHARGE_SHIFT_DESC,
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "69950"
+		})]
+
+	cols += [makeCol("gravy_peptide", "pl7.app/hydrophobicity", "Double",
+		"Hydrophobicity (GRAVY)", dom, {
+			"pl7.app/format": ".3f",
+			"pl7.app/isScore": "true",
+			"pl7.app/score/rankingOrder": "increasing",
+			"pl7.app/description": "rankingOrder: increasing reflects preference for lower hydrophobicity. Invert direction in Lead Selection for hydrophobic-target applications.",
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "69900"
+		})]
+
+	cols += [makeCol("mw_peptide", "pl7.app/molecularWeight", "Double",
+		"Molecular Weight (Da, average masses)", dom, {
+			"pl7.app/format": ".1f",
+			"pl7.app/min": "0",
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "69800"
+		})]
+
+	cols += [makeCol("pi_peptide", "pl7.app/isoelectricPoint", "Double",
+		"Isoelectric Point (pI)", dom, {
+			"pl7.app/format": ".2f",
+			"pl7.app/min": "0",
+			"pl7.app/max": "14",
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "69700"
+		})]
+
+	cols += [makeCol("eox_peptide", "pl7.app/extinctionCoefficientOx", "Double",
+		"Extinction Coeff., Oxidized (M⁻¹cm⁻¹)", dom, {
+			"pl7.app/format": ".0f",
+			"pl7.app/min": "0",
+			"pl7.app/description": "Assumes all Cys are in disulfide bonds. For unprotected linear peptides use the reduced form.",
+			"pl7.app/table/visibility": "optional",
+			"pl7.app/table/orderPriority": "69600"
+		})]
+
+	cols += [makeCol("ered_peptide", "pl7.app/extinctionCoefficientRed", "Double",
+		"Extinction Coeff., Reduced (M⁻¹cm⁻¹)", dom, {
+			"pl7.app/format": ".0f",
+			"pl7.app/min": "0",
+			"pl7.app/description": "Extinction coefficient at 280 nm, disulfide bonds reduced (Cys contribution omitted). A value of 0 means no Tyr or Trp — A280-based quantification is not possible.",
+			"pl7.app/table/visibility": "optional",
+			"pl7.app/table/orderPriority": "69500"
+		})]
+
+	cols += [makeCol("instability_peptide", "pl7.app/instabilityIndex", "Double",
+		"Instability Index", dom, {
+			"pl7.app/format": ".2f",
+			"pl7.app/description": "Guruprasad index — derived from globular proteins. The II > 40 threshold does not apply to short linear peptides; use as a relative composition ranking aid only.",
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "69400"
+		})]
+
+	cols += [makeCol("aliphatic_peptide", "pl7.app/aliphaticIndex", "Double",
+		"Aliphatic Index", dom, {
+			"pl7.app/format": ".1f",
+			"pl7.app/min": "0",
+			"pl7.app/description": "Measures fraction of nonpolar aliphatic residues (Ala, Val, Ile, Leu). For short linear peptides, thermostability interpretation does not apply — the Ikai index was derived for globular mesophilic enzymes, and thermostability is not a meaningful concept for unstructured peptides. Useful as a composition indicator and a proxy for hydrophobic character alongside GRAVY — both metrics increase with Ala/Val/Ile/Leu content, but neither has a universal preferred direction for therapeutic peptides.",
+			"pl7.app/table/visibility": "optional",
+			"pl7.app/table/orderPriority": "69300"
+		})]
+
+	cols += [makeCol("aromaticity_peptide", "pl7.app/aromaticity", "Double",
+		"Aromaticity", dom, {
+			"pl7.app/format": ".3f",
+			"pl7.app/min": "0",
+			"pl7.app/max": "1",
+			"pl7.app/description": "Fraction of aromatic residues (Phe, Trp, Tyr).",
+			"pl7.app/table/visibility": "optional",
+			"pl7.app/table/orderPriority": "69200"
+		})]
+
+	return cols
+}
+
+// CDR-H3 (chain A) and CDR-L3 (chain B) carry different descriptions per
+// pcolumn-spec.md — different developability signals.
+CDR3_CHARGE_DESC := {
+	A: "Strongly positive CDR3 charge correlates with polyreactivity via electrostatic interactions. No universal preferred direction in Lead Selection. IPC 2.0 peptide pKa set.",
+	B: "Strongly positive CDR-L3 charge contributes to paratope polyreactivity. Strongly negative charge is primarily a PK concern. No universal preferred direction. IPC 2.0 peptide pKa set."
+}
+CDR3_GRAVY_DESC := {
+	A: "Lower hydrophobicity preferred for developability. CDR3 GRAVY > 0 is an informal aggregation/polyreactivity heuristic.",
+	B: "Same aggregation and polyreactivity signal as CDR-H3 hydrophobicity; lower independent predictive weight. The TAP score uses combined 6-CDR GRAVY — CDR-L3 alone has limited independent validation."
+}
+
+buildCdr3Columns := func(receptor, chains) {
+	cols := []
+	cdr3OrderA := 68000
+	cdr3OrderB := 67700
+	for chain in chains {
+		frag := labelFragments(receptor, chain)
+		cdr3Dom := { "pl7.app/feature": "CDR3", "pl7.app/vdj/scClonotypeChain": chain }
+		chargeOrder := (chain == "A" ? cdr3OrderA : cdr3OrderB)
+		gravyOrder := chargeOrder - 100
+
+		cols += [makeCol("charge_" + chain + "_CDR3", "pl7.app/charge", "Double",
+			frag.cdr3 + " Net Charge (pH 7)", cdr3Dom, {
+				"pl7.app/format": ".2f",
+				"pl7.app/isScore": "true",
+				"pl7.app/description": CDR3_CHARGE_DESC[chain],
+				"pl7.app/table/visibility": "default",
+				"pl7.app/table/orderPriority": string(chargeOrder)
+			})]
+		cols += [makeCol("chargeShift_" + chain + "_CDR3", "pl7.app/chargeShift", "Double",
+			frag.cdr3 + " ΔCharge (pH 7.4 → 6.0)", {
+				"pl7.app/feature": "CDR3",
+				"pl7.app/vdj/scClonotypeChain": chain,
+				"pl7.app/pH/from": CHARGE_SHIFT_PH_FROM,
+				"pl7.app/pH/to": CHARGE_SHIFT_PH_TO
+			}, {
+				"pl7.app/format": ".2f",
+				"pl7.app/description": CHARGE_SHIFT_DESC,
+				"pl7.app/table/visibility": "default",
+				"pl7.app/table/orderPriority": string(chargeOrder - 50)
+			})]
+		cols += [makeCol("gravy_" + chain + "_CDR3", "pl7.app/hydrophobicity", "Double",
+			frag.cdr3 + " Hydrophobicity (GRAVY)", cdr3Dom, {
+				"pl7.app/format": ".3f",
+				"pl7.app/isScore": "true",
+				"pl7.app/score/rankingOrder": "increasing",
+				"pl7.app/description": CDR3_GRAVY_DESC[chain],
+				"pl7.app/table/visibility": "default",
+				"pl7.app/table/orderPriority": string(gravyOrder)
+			})]
+	}
+	return cols
+}
+
+buildFullChainColumns := func(receptor, fullChains) {
+	cols := []
+	fcOrderBaseA := 67000
+	fcOrderBaseB := 66000
+	for chain in fullChains {
+		frag := labelFragments(receptor, chain)
+		fcDom := { "pl7.app/feature": "VDJRegion", "pl7.app/vdj/scClonotypeChain": chain }
+		base := (chain == "A" ? fcOrderBaseA : fcOrderBaseB)
+		fcLabel := frag.fullChain
+
+		cols += [makeCol("charge_" + chain + "_VDJRegion", "pl7.app/charge", "Double",
+			fcLabel + " Net Charge (pH 7)", fcDom, {
+				"pl7.app/format": ".2f",
+				"pl7.app/isScore": "true",
+				"pl7.app/description": "Non-monotonic vs developability: strongly positive correlates with polyreactivity; strongly negative with rapid clearance.",
+				"pl7.app/table/visibility": "default",
+				"pl7.app/table/orderPriority": string(base)
+			})]
+		cols += [makeCol("pi_" + chain + "_VDJRegion", "pl7.app/isoelectricPoint", "Double",
+			fcLabel + " Isoelectric Point (pI)", fcDom, {
+				"pl7.app/format": ".2f",
+				"pl7.app/isScore": "true",
+				"pl7.app/min": "0",
+				"pl7.app/max": "14",
+				"pl7.app/table/visibility": "default",
+				"pl7.app/table/orderPriority": string(base - 100)
+			})]
+		cols += [makeCol("gravy_" + chain + "_VDJRegion", "pl7.app/hydrophobicity", "Double",
+			fcLabel + " Hydrophobicity (GRAVY)", fcDom, {
+				"pl7.app/format": ".3f",
+				"pl7.app/description": "Framework regions dominate; weak developability signal at chain level — CDR3 hydrophobicity is more discriminating.",
+				"pl7.app/table/visibility": "default",
+				"pl7.app/table/orderPriority": string(base - 200)
+			})]
+		cols += [makeCol("mw_" + chain + "_VDJRegion", "pl7.app/molecularWeight", "Double",
+			fcLabel + " Molecular Weight (Da, average masses)", fcDom, {
+				"pl7.app/format": ".1f",
+				"pl7.app/min": "0",
+				"pl7.app/description": "Unglycosylated sequence mass — does not include N-glycan contributions from any NXS/NXT sequons in the variable region.",
+				"pl7.app/table/visibility": "optional",
+				"pl7.app/table/orderPriority": string(base - 300)
+			})]
+		cols += [makeCol("eox_" + chain + "_VDJRegion", "pl7.app/extinctionCoefficientOx", "Double",
+			fcLabel + " Extinction Coeff., Oxidized (M⁻¹cm⁻¹)", fcDom, {
+				"pl7.app/format": ".0f",
+				"pl7.app/min": "0",
+				"pl7.app/table/visibility": "optional",
+				"pl7.app/table/orderPriority": string(base - 400)
+			})]
+		cols += [makeCol("ered_" + chain + "_VDJRegion", "pl7.app/extinctionCoefficientRed", "Double",
+			fcLabel + " Extinction Coeff., Reduced (M⁻¹cm⁻¹)", fcDom, {
+				"pl7.app/format": ".0f",
+				"pl7.app/min": "0",
+				"pl7.app/table/visibility": "optional",
+				"pl7.app/table/orderPriority": string(base - 500)
+			})]
+		cols += [makeCol("instability_" + chain + "_VDJRegion", "pl7.app/instabilityIndex", "Double",
+			fcLabel + " Instability Index", fcDom, {
+				"pl7.app/format": ".2f",
+				"pl7.app/description": "Guruprasad index, calibrated for in-vitro stability of soluble globular proteins via dipeptide composition. Weak predictor of antibody Tm — use as supplementary ranking aid.",
+				"pl7.app/table/visibility": "optional",
+				"pl7.app/table/orderPriority": string(base - 600)
+			})]
+		cols += [makeCol("aliphatic_" + chain + "_VDJRegion", "pl7.app/aliphaticIndex", "Double",
+			fcLabel + " Aliphatic Index", fcDom, {
+				"pl7.app/format": ".1f",
+				"pl7.app/min": "0",
+				"pl7.app/description": "Ikai aliphatic index, derived from globular mesophilic enzymes. Weak correlation with antibody Tm. No rankingOrder — high values can correlate with aggregation propensity.",
+				"pl7.app/table/visibility": "optional",
+				"pl7.app/table/orderPriority": string(base - 700)
+			})]
+		cols += [makeCol("aromaticity_" + chain + "_VDJRegion", "pl7.app/aromaticity", "Double",
+			fcLabel + " Aromaticity", fcDom, {
+				"pl7.app/format": ".3f",
+				"pl7.app/min": "0",
+				"pl7.app/max": "1",
+				"pl7.app/description": "Fraction of aromatic residues (Phe, Trp, Tyr) over the full chain. Framework dominates; CDR-specific aromaticity is a stronger predictor (Phase 2).",
+				"pl7.app/table/visibility": "optional",
+				"pl7.app/table/orderPriority": string(base - 800)
+			})]
+	}
+	return cols
+}
+
+buildFvColumns := func() {
+	cols := []
+	fvDom := { "pl7.app/feature": "Fv" }
+
+	cols += [makeCol("charge_Fv", "pl7.app/charge", "Double",
+		"Fv Net Charge (pH 7)", fvDom, {
+			"pl7.app/format": ".2f",
+			"pl7.app/isScore": "true",
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "65100"
+		})]
+	cols += [makeCol("chargeShift_Fv", "pl7.app/chargeShift", "Double",
+		"Fv ΔCharge (pH 7.4 → 6.0)", {
+			"pl7.app/feature": "Fv",
+			"pl7.app/pH/from": CHARGE_SHIFT_PH_FROM,
+			"pl7.app/pH/to": CHARGE_SHIFT_PH_TO
+		}, {
+			"pl7.app/format": ".2f",
+			"pl7.app/description": CHARGE_SHIFT_DESC,
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "65050"
+		})]
+	cols += [makeCol("pi_Fv", "pl7.app/isoelectricPoint", "Double",
+		"Fv Isoelectric Point (pI)", fvDom, {
+			"pl7.app/format": ".2f",
+			"pl7.app/isScore": "true",
+			"pl7.app/min": "0",
+			"pl7.app/max": "14",
+			"pl7.app/description": "Variable region (VH+VL) only. Fv pI is typically 2–4 pH units higher than whole-IgG cIEF measurements, which include constant regions (IgG1 Fc pI ≈ 5–6).",
+			"pl7.app/table/visibility": "default",
+			"pl7.app/table/orderPriority": "65000"
+		})]
+	cols += [makeCol("eox_Fv", "pl7.app/extinctionCoefficientOx", "Double",
+		"Fv Extinction Coeff., Oxidized (M⁻¹cm⁻¹)", fvDom, {
+			"pl7.app/format": ".0f",
+			"pl7.app/min": "0",
+			"pl7.app/description": "Variable region (VH+VL) only — does not include constant regions. For whole-IgG A280 quantification, use the full-antibody ε.",
+			"pl7.app/table/visibility": "optional",
+			"pl7.app/table/orderPriority": "64900"
+		})]
+	cols += [makeCol("ered_Fv", "pl7.app/extinctionCoefficientRed", "Double",
+		"Fv Extinction Coeff., Reduced (M⁻¹cm⁻¹)", fvDom, {
+			"pl7.app/format": ".0f",
+			"pl7.app/min": "0",
+			"pl7.app/description": "Variable region (VH+VL) only, disulfide bonds reduced (Cys contribution omitted). A value of 0 means no Tyr or Trp — A280-based quantification is not possible.",
+			"pl7.app/table/visibility": "optional",
+			"pl7.app/table/orderPriority": "64800"
+		})]
+	cols += [makeCol("mw_Fv", "pl7.app/molecularWeight", "Double",
+		"Fv Molecular Weight (Da, average masses)", fvDom, {
+			"pl7.app/format": ".1f",
+			"pl7.app/min": "0",
+			"pl7.app/description": "Unglycosylated sequence mass (VH + VL).",
+			"pl7.app/table/visibility": "optional",
+			"pl7.app/table/orderPriority": "64700"
+		})]
+	return cols
+}
+
+buildColumns := func(args) {
+	if args.mode == "peptide" {
+		return buildPeptideColumns()
+	}
+	cols := buildCdr3Columns(args.receptor, args.chains)
+	cols += buildFullChainColumns(args.receptor, args.fullChains)
+	if args.hasFv {
+		cols += buildFvColumns()
+	}
+	return cols
+}
+
+// ---------------------------------------------------------------------------
+// Spec-cloning helper. Builds a fresh spec dict with optional domain and
+// annotation extras. Used by both getters so the two outputs never share
+// dict references.
+// ---------------------------------------------------------------------------
+
+cloneSpec := func(spec, domainExtras, annotationExtras) {
+	newDomain := {}
+	if spec.domain {
+		for k, v in spec.domain { newDomain[k] = v }
+	}
+	if domainExtras {
+		for k, v in domainExtras { newDomain[k] = v }
+	}
+	newAnnotations := {}
+	if spec.annotations {
+		for k, v in spec.annotations { newAnnotations[k] = v }
+	}
+	if annotationExtras {
+		for k, v in annotationExtras { newAnnotations[k] = v }
+	}
+	return {
+		name: spec.name,
+		valueType: spec.valueType,
+		domain: newDomain,
+		annotations: newAnnotations
+	}
+}
+
+wrap := func(col, newSpec) {
+	return {
+		column: col.column,
+		id: col.id,
+		naRegex: col.naRegex,
+		allowNA: col.allowNA,
+		spec: newSpec
+	}
+}
+
+// ---------------------------------------------------------------------------
+// Public getters.
+// ---------------------------------------------------------------------------
+
+forPropertiesPf := func(args) {
+	cols := buildColumns(args)
+	out := []
+	for col in cols {
+		out += [wrap(col, cloneSpec(col.spec, undefined, { "pl7.app/isOutput": "true" }))]
+	}
+	return out
+}
+
+forExport := func(args, blockId) {
+	cols := buildColumns(args)
+	out := []
+	for col in cols {
+		if !col.spec.annotations { continue }
+		if col.spec.annotations["pl7.app/isScore"] != "true" { continue }
+		out += [wrap(col, cloneSpec(col.spec, { "pl7.app/blockId": blockId }, undefined))]
+	}
+	return out
+}
+
+export ll.toStrict({
+	forPropertiesPf: forPropertiesPf,
+	forExport: forExport
+})

package/src/messages.lib.tengo

@@ -47,6 +47,12 @@ vhh := func() {
 		"disregard these thresholds for nanobody libraries."
 }
 
+// R9 — Guruprasad instability index is undefined for sequences shorter than
+// 10 residues; emitted in peptide mode whenever any peptide falls below the floor.
+peptidesShortInstability := func() {
+	return "Instability Index applies only to peptides ≥10 aa; shorter peptides show blank values."
+}
+
 export ll.toStrict({
 	partialChainMissingFullChain: partialChainMissingFullChain,
 	partialChainNoCdr3: partialChainNoCdr3,
@@ -54,5 +60,6 @@ export ll.toStrict({
 	cdr3OnlyInput: cdr3OnlyInput,
 	gammaDeltaTcr: gammaDeltaTcr,
 	receptorNotDetected: receptorNotDetected,
-	vhh: vhh
+	vhh: vhh,
+	peptidesShortInstability: peptidesShortInstability
 })