Fixed parameters and error in boltz2 message printing.

Author: FloWuenne

assets/schema_input_design.json

@@ -52,8 +52,17 @@
       "target_sequence": {
         "type": "string",
         "errorMessage": "Target sequence must be a valid file path to a FASTA file containing the target protein sequence"
+      },
+      "target_template": {
+        "type": "string",
+        "pattern": "^\\S+\\.cif$",
+        "errorMessage": "Target template must be a valid file path to a CIF file (e.g., 'target_structure.cif')"
       }
     },
-    "required": ["sample_id", "design_yaml", "target_sequence"]
+    "required": [
+      "sample_id",
+      "design_yaml",
+      "target_sequence"
+    ]
   }
-}
+}

assets/test_data/samplesheet_design_nanobody.csv

@@ -1,2 +1,2 @@
-sample_id,design_yaml,structure_files,protocol,num_designs,budget,reuse,target_msa,target_sequence
-design1_nano,assets/test_data/nipah_nanobody_design.yaml,assets/test_data/nipah_virus_Glycoprotein_competition_structure.cif,nanobody-anything,3,2,,assets/test_data/nipah_glycoprotein_msa_Uniref30_2302.a3m,assets/test_data/nipah_virus_target_sequence_glycoproteinG.fasta
+sample_id,design_yaml,structure_files,protocol,num_designs,budget,reuse,target_msa,target_sequence,target_template
+design1_nano,assets/test_data/nipah_nanobody_design.yaml,assets/test_data/nipah_virus_Glycoprotein_competition_structure.cif,nanobody-anything,3,2,,assets/test_data/nipah_glycoprotein_msa_Uniref30_2302.a3m,assets/test_data/nipah_virus_target_sequence_glycoproteinG.fasta,assets/test_data/nipah_virus_Glycoprotein_competition_structure.cif

assets/test_data/samplesheet_design_peptide.csv

@@ -1,2 +1,2 @@
-sample_id,design_yaml,structure_files,protocol,num_designs,budget,reuse,target_msa,target_sequence
-design1_pep,assets/test_data/nipah_peptide_design.yaml,assets/test_data/nipah_virus_Glycoprotein_competition_structure.cif,peptide-anything,3,2,,,assets/test_data/2VSM_target_sequence.fa
+sample_id,design_yaml,structure_files,protocol,num_designs,budget,reuse,target_msa,target_sequence,target_template
+design1_pep,assets/test_data/nipah_peptide_design.yaml,assets/test_data/nipah_virus_Glycoprotein_competition_structure.cif,peptide-anything,3,2,,,assets/test_data/2VSM_target_sequence.fa,

assets/test_data/samplesheet_design_protein.csv

@@ -1,2 +1,2 @@
-sample_id,design_yaml,structure_files,protocol,num_designs,budget,reuse,target_msa,target_sequence
-design1_prot,assets/test_data/nipah_protein_design.yaml,assets/test_data/nipah_virus_Glycoprotein_competition_structure.cif,protein-anything,3,2,,assets/test_data/nipah_glycoprotein_msa_Uniref30_2302.a3m,assets/test_data/nipah_virus_target_sequence_glycoproteinG.fasta
+sample_id,design_yaml,structure_files,protocol,num_designs,budget,reuse,target_msa,target_sequence,target_template
+design1_prot,assets/test_data/nipah_protein_design.yaml,assets/test_data/nipah_virus_Glycoprotein_competition_structure.cif,protein-anything,3,2,,assets/test_data/nipah_glycoprotein_msa_Uniref30_2302.a3m,assets/test_data/nipah_virus_target_sequence_glycoproteinG.fasta,

bin/prepare_boltz2_input.py

@@ -10,6 +10,7 @@ def parse_args():
     parser.add_argument('--mpnn_sequences', required=True, help='Path to ProteinMPNN sequences FASTA file')
     parser.add_argument('--target_sequence', required=True, help='Target sequence string')
     parser.add_argument('--target_msa', help='Path to target MSA file')
+    parser.add_argument('--target_template', help='Path to target template CIF file')
     parser.add_argument('--meta_id', required=True, help='ID for the current design')
     parser.add_argument('--parent_id', required=True, help='Parent ID')
     parser.add_argument('--predict_affinity', action='store_true', help='Enable affinity prediction')
@@ -34,6 +35,13 @@ def main():
         has_target_msa = True
         target_msa_path = args.target_msa
 
+    # Check if target template is provided and valid
+    has_target_template = False
+    target_template_path = None
+    if args.target_template and args.target_template != 'NO_TEMPLATE' and os.path.exists(args.target_template):
+        has_target_template = True
+        target_template_path = args.target_template
+    
     os.makedirs(args.output_dir, exist_ok=True)
 
     # Process each FASTA file
@@ -104,6 +112,18 @@ def main():
             if has_target_msa and target_msa_path:
                 target_entry['protein']['msa'] = os.path.abspath(target_msa_path)
                 print(f"    Adding target MSA: {target_msa_path}")
+            
+            # Add template configuration if provided
+            # Template is always applied to chain B (target) with force=true and threshold=1.0
+            if has_target_template and target_template_path:
+                target_entry['protein']['templates'] = [{
+                    'cif': os.path.abspath(target_template_path),
+                    'force': True,
+                    'threshold': 1.0,
+                    'chain_id': 'B'
+                }]
+                print(f"    Adding target template: {target_template_path} (force=true, threshold=1.0)")
+            
             # Build final YAML input with exactly two entries
             boltz2_input = {
                 'version': 1,
@@ -127,7 +147,7 @@ def main():
                 yaml.dump(boltz2_input, yf, default_flow_style=False)
 
             print(f"  Created YAML input: {yaml_file}")
-            print(f"    Binder length: {len(binder_seq)}")
+            print(f"    Binder length: {len(binder_seq_clean)}")
             print(f"    Target length: {len(target_seq)}")
             print(f"    Target MSA: {'Yes' if has_target_msa else 'No (will use sequence only)'}")
             print(f"    Binder MSA: Inferred automatically by Boltz-2")

main.nf

@@ -106,7 +106,7 @@ workflow NFPROTEINDESIGN {
         .fromList(design_samplesheet)
         .map { tuple ->
             // samplesheetToList returns list of values in schema order
-            // Order: sample_id, design_yaml, structure_files, protocol, num_designs, budget, reuse, target_msa, target_sequence
+            // Order: sample_id, design_yaml, structure_files, protocol, num_designs, budget, reuse, target_msa, target_sequence, target_template
             def sample_id = tuple[0]
             def design_yaml_path = tuple[1]
             def structure_files_str = tuple[2]
@@ -116,6 +116,7 @@ workflow NFPROTEINDESIGN {
             def reuse = tuple.size() > 6 ? tuple[6] : null
             def target_msa_path = tuple.size() > 7 ? tuple[7] : null
             def target_sequence_path = tuple.size() > 8 ? tuple[8] : null
+            def target_template_path = tuple.size() > 9 ? tuple[9] : null
 
             // Convert design YAML to file object and validate existence
             // Smart path resolution: try launchDir first (for local runs), then projectDir (for Platform)
@@ -182,14 +183,29 @@ workflow NFPROTEINDESIGN {
                 }
             }
 
+            // Parse target template CIF file (optional for Boltz2 refolding)
+            def target_template = null
+            if (target_template_path) {
+                if (target_template_path.startsWith('/') || target_template_path.contains('://')) {
+                    target_template = file(target_template_path, checkIfExists: true)
+                } else {
+                    def launchDir_path = file(target_template_path)
+                    if (launchDir_path.exists()) {
+                        target_template = launchDir_path
+                    } else {
+                        target_template = file("${project_dir}/${target_template_path}", checkIfExists: true)
+                    }
+                }
+            }
+
             def meta = [:]
             meta.id = sample_id
             meta.protocol = protocol ?: params.protocol
             meta.num_designs = num_designs ?: params.num_designs
             meta.budget = budget ?: params.budget
             meta.reuse = reuse ?: false
 
-            [meta, design_yaml, structure_files, target_msa, target_sequence]
+            [meta, design_yaml, structure_files, target_msa, target_sequence, target_template]
         }
 
     // ========================================================================

modules/local/boltz2_refold.nf

@@ -27,7 +27,7 @@ process BOLTZ2_REFOLD {
     accelerator 1, type: 'nvidia-gpu'
 
     input:
-    tuple val(meta), path(mpnn_sequences), path(target_sequence_file), path(target_msa)
+    tuple val(meta), path(mpnn_sequences), path(target_sequence_file), path(target_msa), path(target_template)
     path cache_dir
 
     output:
@@ -44,6 +44,7 @@ process BOLTZ2_REFOLD {
     def num_recycling = params.boltz2_num_recycling ?: 3
     def num_diffusion = params.boltz2_num_diffusion ?: 5
     def has_target_msa = target_msa.name != 'NO_MSA'
+    def has_target_template = target_template.name != 'NO_TEMPLATE'
     """
     #!/bin/bash
     set -euo pipefail
@@ -89,6 +90,7 @@ process BOLTZ2_REFOLD {
         --mpnn_sequences "${mpnn_sequences}" \\
         --target_sequence "\$TARGET_SEQ" \\
         --target_msa "${target_msa}" \\
+        --target_template "${target_template}" \\
         --meta_id "${meta.id}" \\
         --parent_id "${meta.parent_id}" \\
         --output_dir "yaml_inputs" \\

nextflow.config

@@ -20,16 +20,14 @@ params {
     // ========================================================================
     // Design and Boltzgen parameters
     // ========================================================================
-    // IMPORTANT: The following parameters should be specified in your samplesheet:
+    // IMPORTANT: The following parameters must be specified in your samplesheet:
     //   - protocol: Boltzgen protocol (protein-anything, peptide-anything, etc.)
     //   - num_designs: Number of intermediate designs to generate
     //   - budget: Number of designs in final diversity-optimized set
     //
     // This design ensures explicit per-sample control and eliminates ambiguity.
     // See samplesheet schema and examples in assets/test_data/ for details.
     // ========================================================================
-    num_designs                = null                // Number of intermediate designs (typically specified per-sample in samplesheet)
-    budget                     = null                // Number of final budget designs (typically specified per-sample in samplesheet)
 
     // Boltzgen advanced options
     cache_dir                  = null                // Cache directory for model weights (~6GB), defaults to ~/.cache
@@ -41,7 +39,7 @@ params {
     mpnn_sampling_temp         = 0.1                 // Sampling temperature (0.1-0.3 recommended, lower = more conservative)
     mpnn_num_seq_per_target    = 8                   // Number of sequence variants to generate per structure
     mpnn_batch_size            = 1                   // Batch size for ProteinMPNN inference
-    mpnn_seed                  = 37                  // Random seed for reproducibility
+    mpnn_seed                  = 37                  // Random seed for reproducibility 
     mpnn_backbone_noise        = 0.02                // Backbone noise level (0.02-0.20, lower = more faithful to input)
     mpnn_save_score            = true                // Save per-residue scores
     mpnn_save_probs            = false               // Save per-residue probabilities (large files, use for detailed analysis)

nextflow_schema.json

@@ -22,7 +22,7 @@
                     "mimetype": "text/csv",
                     "pattern": "^\\S+\\.csv$",
                     "description": "Path to comma-separated samplesheet file.",
-                    "help_text": "The samplesheet must contain: `sample`, `design_yaml`, and optionally: `structure_files`, `protocol`, `num_designs`, `budget`, `reuse`\n\nSee schema file in assets/schema_input_design.json for detailed specifications.",
+                    "help_text": "The samplesheet must contain: `sample_id`, `design_yaml`, `target_sequence`, and optionally: `structure_files`, `protocol`, `num_designs`, `budget`, `reuse`, `target_msa`, `target_template`\n\nSee schema file in assets/schema_input_design.json for detailed specifications.",
                     "fa_icon": "fas fa-file-csv"
                 },
                 "outdir": {
@@ -41,35 +41,6 @@
             "default": "",
             "fa_icon": "fas fa-cogs",
             "properties": {
-                "protocol": {
-                    "type": "string",
-                    "default": "protein-anything",
-                    "description": "Boltzgen design protocol.",
-                    "help_text": "Available protocols:\n- **protein-anything**: General protein design\n- **peptide-anything**: Peptide binder design\n- **protein-small_molecule**: Protein-small molecule interaction\n- **nanobody-anything**: Nanobody design",
-                    "enum": [
-                        "protein-anything",
-                        "peptide-anything",
-                        "protein-small_molecule",
-                        "nanobody-anything"
-                    ],
-                    "fa_icon": "fas fa-flask"
-                },
-                "num_designs": {
-                    "type": "integer",
-                    "default": 100,
-                    "description": "Number of intermediate designs to generate.",
-                    "help_text": "For production runs, recommend 10,000-60,000 designs. Lower values (100-1000) suitable for testing.",
-                    "fa_icon": "fas fa-hashtag",
-                    "minimum": 1
-                },
-                "budget": {
-                    "type": "integer",
-                    "default": 10,
-                    "description": "Number of designs in final diversity-optimized set.",
-                    "help_text": "Boltzgen will select this many diverse, high-quality designs from the intermediate pool.",
-                    "fa_icon": "fas fa-star",
-                    "minimum": 1
-                },
                 "cache_dir": {
                     "type": "string",
                     "format": "directory-path",

workflows/protein_design.nf

@@ -19,7 +19,7 @@ include { CONSOLIDATE_METRICS } from '../modules/local/consolidate_metrics'
 workflow PROTEIN_DESIGN {
 
     take:
-    ch_input         // channel: [meta, design_yaml, structure_files, target_msa, target_sequence]
+    ch_input         // channel: [meta, design_yaml, structure_files, target_msa, target_sequence, target_template]
     ch_cache         // channel: path to cache directory or EMPTY_CACHE placeholder
     ch_boltz2_cache  // channel: path to Boltz-2 cache directory or EMPTY_BOLTZ2_CACHE placeholder
 
@@ -29,9 +29,9 @@ workflow PROTEIN_DESIGN {
     // Run Boltzgen on design YAMLs
     // ========================================================================
 
-    // Prepare Boltzgen input by removing target_msa and target_sequence (not needed for Boltzgen)
+    // Prepare Boltzgen input by removing target_msa, target_sequence, and target_template (not needed for Boltzgen)
     ch_boltzgen_input = ch_input
-        .map { meta, design_yaml, structure_files, target_msa, target_sequence ->
+        .map { meta, design_yaml, structure_files, target_msa, target_sequence, target_template ->
             [meta, design_yaml, structure_files]
         }
 
@@ -86,7 +86,7 @@ workflow PROTEIN_DESIGN {
         if (params.run_boltz2_refold) {
             // Get target sequence FASTA from samplesheet
             ch_target_fasta = ch_input
-                .map { meta, design_yaml, structure_files, target_msa, target_sequence ->
+                .map { meta, design_yaml, structure_files, target_msa, target_sequence, target_template ->
                     [meta.id, target_sequence]
                 }
 
@@ -113,11 +113,20 @@ workflow PROTEIN_DESIGN {
             // Prepare Target MSA from Samplesheet
             // ================================================================
             ch_target_msa = ch_input
-                .map { meta, design_yaml, structure_files, target_msa, target_sequence ->
+                .map { meta, design_yaml, structure_files, target_msa, target_sequence, target_template ->
                     def msa_file = target_msa ?: file('NO_MSA')
                     [meta.id, msa_file]
                 }
 
+            // ================================================================
+            // Prepare Target Template from Samplesheet
+            // ================================================================
+            ch_target_template = ch_input
+                .map { meta, design_yaml, structure_files, target_msa, target_sequence, target_template ->
+                    def template_file = target_template ?: file('NO_TEMPLATE')
+                    [meta.id, template_file]
+                }
+
             // ================================================================
             // Create channel for Boltz-2 refolding
             // ================================================================
@@ -156,7 +165,11 @@ workflow PROTEIN_DESIGN {
                 }
                 .combine(ch_target_msa, by: 0)
                 .map { parent_id, meta, fasta, target_seq, target_msa ->
-                    [meta, fasta, target_seq, target_msa]
+                    [meta.parent_id, meta, fasta, target_seq, target_msa]
+                }
+                .combine(ch_target_template, by: 0)
+                .map { parent_id, meta, fasta, target_seq, target_msa, target_template ->
+                    [meta, fasta, target_seq, target_msa, target_template]
                 }
 
             // Run Boltz-2 structure prediction with target MSA