Add LigandMPNN Nextflow pipeline for protein sequence design

params.json

@@ -0,0 +1,131 @@
+{
+    "params": {
+        "pdb": {
+            "type": "file",
+            "description": "Path to input PDB file for protein sequence design",
+            "default": "/mnt/OmicNAS/private/old/olamide/ligandmpnn/input/1BC8.pdb",
+            "required": true,
+            "pipeline_io": "input",
+            "var_name": "params.pdb",
+            "examples": [
+                "/mnt/workflow/input/protein.pdb",
+                "/mnt/workflow/input/*.pdb"
+            ],
+            "pattern": ".*\\.pdb$",
+            "enum": [],
+            "validation": {},
+            "notes": "Input PDB file containing the protein structure for sequence design."
+        },
+        "outdir": {
+            "type": "folder",
+            "description": "Directory for LigandMPNN output results",
+            "default": "/mnt/OmicNAS/private/old/olamide/ligandmpnn/output",
+            "required": true,
+            "pipeline_io": "output",
+            "var_name": "params.outdir",
+            "examples": [
+                "/mnt/workflow/output",
+                "/path/to/results"
+            ],
+            "pattern": ".*",
+            "enum": [],
+            "validation": {},
+            "notes": "Directory where designed sequences and backbone PDBs will be saved."
+        },
+        "model_type": {
+            "type": "string",
+            "description": "Type of MPNN model to use for sequence design",
+            "default": "ligand_mpnn",
+            "required": false,
+            "pipeline_io": "parameter",
+            "var_name": "params.model_type",
+            "examples": [
+                "protein_mpnn",
+                "ligand_mpnn",
+                "soluble_mpnn"
+            ],
+            "pattern": "^(protein_mpnn|ligand_mpnn|soluble_mpnn|global_label_membrane_mpnn|per_residue_label_membrane_mpnn)$",
+            "enum": ["protein_mpnn", "ligand_mpnn", "soluble_mpnn", "global_label_membrane_mpnn", "per_residue_label_membrane_mpnn"],
+            "validation": {},
+            "notes": "protein_mpnn: Original ProteinMPNN. ligand_mpnn: Context-aware with ligands. soluble_mpnn: Trained on soluble proteins."
+        },
+        "temperature": {
+            "type": "number",
+            "description": "Sampling temperature for sequence generation",
+            "default": 0.1,
+            "required": false,
+            "pipeline_io": "parameter",
+            "var_name": "params.temperature",
+            "examples": [0.05, 0.1, 0.2],
+            "pattern": null,
+            "enum": [],
+            "validation": {},
+            "notes": "Higher temperature gives more sequence diversity. Recommended range: 0.05-0.5"
+        },
+        "seed": {
+            "type": "integer",
+            "description": "Random seed for reproducibility",
+            "default": 111,
+            "required": false,
+            "pipeline_io": "parameter",
+            "var_name": "params.seed",
+            "examples": [111, 42, 12345],
+            "pattern": null,
+            "enum": [],
+            "validation": {},
+            "notes": "Set for reproducible results."
+        },
+        "batch_size": {
+            "type": "integer",
+            "description": "Number of sequences to generate per batch",
+            "default": 1,
+            "required": false,
+            "pipeline_io": "parameter",
+            "var_name": "params.batch_size",
+            "examples": [1, 3, 5],
+            "pattern": null,
+            "enum": [],
+            "validation": {},
+            "notes": "Higher batch sizes require more GPU memory."
+        },
+        "number_of_batches": {
+            "type": "integer",
+            "description": "Number of batches to run",
+            "default": 1,
+            "required": false,
+            "pipeline_io": "parameter",
+            "var_name": "params.number_of_batches",
+            "examples": [1, 5, 10],
+            "pattern": null,
+            "enum": [],
+            "validation": {},
+            "notes": "Total sequences = batch_size × number_of_batches"
+        },
+        "chains_to_design": {
+            "type": "string",
+            "description": "Comma-separated chain IDs to redesign",
+            "default": "",
+            "required": false,
+            "pipeline_io": "parameter",
+            "var_name": "params.chains_to_design",
+            "examples": ["A", "A,B", "A,B,C"],
+            "pattern": "^([A-Z],?)*$",
+            "enum": [],
+            "validation": {},
+            "notes": "Leave empty to design all chains."
+        },
+        "fixed_residues": {
+            "type": "string",
+            "description": "Space-separated list of residues to keep fixed",
+            "default": "",
+            "required": false,
+            "pipeline_io": "parameter",
+            "var_name": "params.fixed_residues",
+            "examples": ["A1 A2 A3", "A12 B25 B26"],
+            "pattern": null,
+            "enum": [],
+            "validation": {},
+            "notes": "Format: ChainResidue (e.g., A12). Leave empty to design all residues."
+        }
+    }
+}