covalent-design

Version Compatibility

Reference examples tested with: RDKit 2024.09+, OpenEye / AutoDock Vina 1.2+ (for covalent extensions), GOLD (commercial), DOCKovalent (web service), HCovDock 1.0+.

Before using code patterns, verify installed versions match. If versions differ:

• Python: pip show <package> then help(rdkit.Chem) to check signatures

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Covalent Inhibitor Design

Hard rules

1. No fabricated citations. Every cited work must resolve to a verifiable
2. No claim without provenance. Every quantitative or factual claim
3. No silent failure. Every script invocation, API call, or tool use must declare its exit status and what to do on non-zero. A skill that silently swallows errors is a violation.

When to use

Load this skill when the user asks a question that matches its declared trigger conditions (see the frontmatter description for the most common ones). Do not load it for questions outside its scope — defer to the appropriate ORS skill instead.

When NOT to use

Do not load this skill if the question is in a sibling skill's domain (see ## Cross-references), if the user explicitly asks for a different tool, or if the task is outside the skill's declared category.

Reactive Residue Taxonomy

Cysteine is the dominant target because:

• Soft nucleophile (matches soft electrophiles)
• Low background reactivity (rare in proteins, ~1.7%)
• Distinguishable from common nucleophiles (GSH, off-target Cys)

Warhead Chemistry

Practical hierarchy: Acrylamide is the modern default for cysteine-selective TCIs. Chloroacetamide is more reactive (faster) but less selective.

Decision Tree by Scenario

Intrinsic Reactivity Assays

A covalent inhibitor's effectiveness = (intrinsic reactivity) × (target residence time in pocket) / (off-target reactivity).

WARHEAD_PATTERNS = {
    'acrylamide': Chem.MolFromSmarts('[CX3]=[CX3][CX3]=O'),
    'chloroacetamide': Chem.MolFromSmarts('[CX3](=O)C[Cl]'),
    'vinyl_sulfone': Chem.MolFromSmarts('[SX4](=O)(=O)C=C'),
    'sulfonyl_fluoride': Chem.MolFromSmarts('[SX4](=O)(=O)F'),
}

def warhead_reactivity(mol):
    hits = []
    for name, pat in WARHEAD_PATTERNS.items():
        if mol.HasSubstructMatch(pat):
            hits.append(name)
    return hits

GSH Stability

Test the warhead's intrinsic reactivity against glutathione (γ-Glu-Cys-Gly) in silico before synthesis. A compound that reacts quickly with GSH reacts quickly with off-target thiols (Cys in other proteins).

GSH_SMILES = 'NC(CCC(=O)NCC(=O)O)C(=O)NCC(=O)O'
GSH = Chem.MolFromSmiles(GSH_SMILES)
GSH_CYS_S = [a.GetIdx() for a in GSH.GetAtoms() if a.GetSymbol() == 'S']

# Check warhead accessibility to Cys-SH in GSH
# Use docking or compute GSH-Warhead clash as proxy

For experimental GSH stability: HPLC-MS with GSH (1 mM) + compound (10 µM) at 37°C; measure t1/2 (high-quality TCIs have GSH t1/2 > 4 hours).

Reversible vs Irreversible Covalent

Reversible covalent is preferred when:

• Off-target Cys in related proteins is concern
• Long-term dosing required
• Pharmacodynamic response must be reversible

Covalent Docking Tools

Standard Vina/GNINA cannot predict covalent adducts. Covalent-specific tools:

Workflow: (1) Generate covalent adduct SMILES from warhead + target residue, (2) Build 3D structure, (3) Dock with covalent constraint.

# Conceptual covalent adduct generation
# Use RDKit reaction SMARTS to attach warhead to Cys-S-H
adduct_rxn = AllChem.ReactionFromSmarts(
    '[Cys-S-H:1].[CX3:2]=[CX3:3][CX3:4]=O>>[Cys-S:1][CX3:2][CX3:3][CX3:4]=O'
)

Per-Tool Failure Modes

Acrylamide too reactive in vivo

Trigger: Compound shows GSH t1/2 < 1 hour.

Mechanism: Acrylamide warhead reacts with off-target Cys.

Symptom: Off-target adducts, toxicity.

Fix: Reduce warhead reactivity (alpha-substituted acrylamide, e.g. KRAS G12C strategy); use Cyanoacrylamide for reversible covalent.

No covalent adduct formed in MS

Trigger: Compound has warhead, binds pocket, but no covalent adduct observed.

Mechanism: Warhead geometry wrong; nucleophile not in attack position.

Symptom: Reversible-only binding; no time-dependent IC50 shift.

Fix: Visualize docked pose; check distance nucleophile to electrophile < 4 Å; align with target residue.

Hook effect in PROTAC context

Trigger: PROTAC with covalent target ligand shows bell-shaped degradation curve.

Mechanism: At high PROTAC concentration, binary complexes (target-PROTAC, E3-PROTAC) dominate over ternary.

Symptom: Dmax falls at high [PROTAC]; no degradation.

Fix: Use lower dose window; optimize linker to favor ternary.

References

• Lonsdale & Ward, J. Med. Chem. -- targeted covalent inhibitors review.
• Singh et al. -- KRAS G12C and covalent drug design.
• Gehringer & Laufer, J. Med. Chem. -- cysteine-targeted warheads.

Related Skills

• chemoinformatics/substructure-search - Warhead SMARTS
• chemoinformatics/virtual-screening - Non-covalent docking
• chemoinformatics/protac-degraders - Bifunctional covalent
• chemoinformatics/pose-validation - Pose QC for covalent adducts

Cross-references

Other skills in this category:

• admet-prediction
• conformer-generation
• docking-rescoring
• free-energy-calculations
• generative-design
• molecular-descriptors
• molecular-io
• molecular-standardization
• pharmacophore-modeling
• pose-validation
• protac-degraders
• qsar-modeling
• retrosynthesis
• scaffold-analysis
• shape-similarity
• similarity-searching
• substructure-search
• virtual-screening

Changelog

• 1.1.0 (migration) — Bulk-migrated to v0.4.0 schema: canonical metadata block, base Hard rules, Cross-references. Body content unchanged; author should review and fill in any domain-specific extensions to the Hard rules.
• 1.0.0 — Initial release.