{"skill":{"slug":"pharmaclaw-catalyst-design","displayName":"Pharmaclaw Catalyst Design","summary":"Organometallic catalyst recommendation and novel ligand design for drug synthesis reactions. Recommends catalysts (Pd, Ru, Rh, Ir, Ni, Cu, Zr, Fe) for reacti...","description":"---\nname: pharmaclaw-catalyst-design\ndescription: Organometallic catalyst recommendation and novel ligand design for drug synthesis reactions. Recommends catalysts (Pd, Ru, Rh, Ir, Ni, Cu, Zr, Fe) for reaction types (Suzuki, Heck, Buchwald-Hartwig, metathesis, hydrogenation, click, etc.) from curated database with scoring. Designs novel ligand variants via RDKit (steric, electronic, bioisosteric modifications). Chains from chemistry-query/retrosynthesis (receives reaction type + substrate) and feeds into IP Expansion (novel ligands as patentable inventions). Triggers on catalyst, ligand, organometallic, cross-coupling catalyst, reaction conditions, catalyst selection, ligand design, cone angle, bite angle, phosphine, NHC, palladium catalyst, ruthenium catalyst.\n---\n\n# Catalyst Design Agent v1.0.0\n\n## Overview\n\nRecommends organometallic catalysts for drug synthesis steps and designs novel ligand modifications. Two core workflows: **recommend** (find the right catalyst) and **design** (create novel ligand variants).\n\n## Quick Start\n\n```bash\n# Recommend catalysts for a Suzuki coupling\npython scripts/catalyst_recommend.py --reaction suzuki\n\n# Recommend with constraints (prefer cheap, earth-abundant)\npython scripts/catalyst_recommend.py --reaction \"C-N coupling\" --constraints '{\"prefer_earth_abundant\": true, \"max_cost\": \"medium\"}'\n\n# Design novel ligand variants from PPh3\npython scripts/ligand_designer.py --scaffold PPh3 --strategy all --draw\n\n# Full chain: reaction → catalyst → ligand optimization\npython scripts/chain_entry.py --input-json '{\"reaction\": \"suzuki\", \"context\": \"retrosynthesis\"}'\n```\n\n## Scripts\n\n### `scripts/catalyst_recommend.py`\nScores and ranks catalysts from curated database (12 catalysts, 28 reaction types).\n\n```\n--reaction <type>              Required. e.g., suzuki, metathesis, C-N coupling, hydrogenation\n--substrate <SMILES>           Optional. Substrate context\n--constraints <JSON>           Optional. {prefer_metal, max_cost, prefer_earth_abundant}\n--enantioselective             Flag. Prioritize chiral catalysts\n```\n\nScoring (0-100): reaction match (50), cost (15), metal preference (10), enantioselectivity (10), loading efficiency (5), advantages (5), earth-abundance (5).\n\n### `scripts/ligand_designer.py`\nGenerates novel ligand variants via three strategies:\n\n| Strategy | Method | Output |\n|----------|--------|--------|\n| steric | Add methyl/iPr/tBu to aromatic rings | Modified SMILES + properties |\n| electronic | Add OMe/F/CF3 substituents | Modified SMILES + properties |\n| bioisosteric | P→NHC, phenyl→pyridyl, phosphine→phosphite | Conceptual suggestions + rationale |\n\n```\n--scaffold <SMILES|name>       Required. PPh3, NHC_IMes, NHC_IPr, PCy3, dppe, dppp, or raw SMILES\n--strategy <type>              steric|electronic|bioisosteric|all (default: all)\n--draw                         Generate 2D grid PNG of variants\n--output <path>                Save JSON results to file\n```\n\n### `scripts/chain_entry.py`\nStandard PharmaClaw chain interface. Accepts JSON, routes to recommend/design/both.\n\nInput keys: `reaction`, `scaffold`/`ligand`, `substrate`/`smiles`, `constraints`, `enantioselective`, `strategy`, `draw`, `context`\n\nIf only `reaction` is given, also auto-runs ligand optimization on the top recommended catalyst's ligand.\n\n## Chaining\n\n| From | Input | To | Output |\n|------|-------|----|--------|\n| Chemistry Query / Retrosynthesis | Reaction type needed for a synthesis step | **Catalyst Design** | Ranked catalysts + conditions |\n| **Catalyst Design** | Top catalyst ligand SMILES | Ligand Designer | Novel ligand variants |\n| **Catalyst Design** | Novel ligand SMILES | IP Expansion | Patent landscape check |\n| **Catalyst Design** | Recommended conditions | Chemistry Query | Forward reaction simulation |\n\n## Database\n\n`references/catalyst_database.json` — 12 catalysts, 8 metals, 28 reaction types. Includes SMILES, conditions, loading ranges, cost ratings, advantages/limitations, and literature DOIs.\n\nExpandable: add entries following the existing schema.\n\n## Ligand Aliases\n\nPPh3, PCy3, dppe, dppp, NHC_IMes, NHC_IPr — resolved automatically to SMILES.\n","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":622,"installsAllTime":2,"installsCurrent":2,"stars":0,"versions":1},"createdAt":1773252895650,"updatedAt":1778491839794},"latestVersion":{"version":"1.0.0","createdAt":1773252895650,"changelog":"Initial release of Catalyst Design Agent for organometallic catalyst recommendation and ligand design.\n\n- Recommends catalysts (Pd, Ru, Rh, Ir, Ni, Cu, Zr, Fe) for 28 common reaction types with scoring based on reaction fit, cost, metallic abundance, and more.\n- Designs novel ligand variants using steric, electronic, and bioisosteric modifications via RDKit.\n- Provides scripts for catalyst recommendation, ligand design, and full chemistry workflow chaining.\n- Integrates with retrosynthesis and IP Expansion skills for automated ligand optimization and patentability checks.\n- Curated catalyst database with ligand/condition data and automatic ligand alias resolution.","license":"MIT-0"},"metadata":null,"owner":{"handle":"cheminem","userId":"s17eyy18e8m8evx3p0e0n47c99884hqe","displayName":"Cheminem","image":"https://avatars.githubusercontent.com/u/260304734?v=4"},"moderation":null}