Bio Substructure Search

Version Compatibility

Reference examples tested with: RDKit 2024.09+. SMARTS dialect follows Daylight specification with RDKit extensions.

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

• Python: pip show rdkit then help(rdkit.Chem.MolFromSmarts) 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.

Substructure Search

Search molecular collections for structural patterns using SMARTS. The choice of SMARTS dialect, atom/bond matching mode, and structural-alert catalog determines whether the search is correctly capturing the intended chemistry. PAINS (Baell & Holloway 2010) is the most-cited but most-misunderstood filter -- it identifies patterns of assay interference, not "bad molecules". Knowing when to apply each catalog and how to interpret hits is essential.

For SMARTS-based reactions (transforming matched substructures), see chemoinformatics/reaction-enumeration. For 3D pharmacophore matching, see chemoinformatics/pharmacophore-modeling.

SMARTS Grammar Essentials

Common SMARTS Patterns

Basic Substructure Match

Goal: Test whether a molecule contains a SMARTS pattern and enumerate the matching atom indices.

Approach: Parse the molecule with MolFromSmiles and the pattern with MolFromSmarts, gate with HasSubstructMatch, then call GetSubstructMatches and map each atom index back to the molecule for inspection.

from rdkit import Chem

mol = Chem.MolFromSmiles('c1ccc(O)cc1CCO')
pattern = Chem.MolFromSmarts('[OX2H]')

if mol.HasSubstructMatch(pattern):
    matches = mol.GetSubstructMatches(pattern)
    for match in matches:
        atoms = [mol.GetAtomWithIdx(i).GetSymbol() for i in match]

HasSubstructMatch returns bool, GetSubstructMatches returns tuple of tuples of atom indices.

Recursive SMARTS (key for postdoc-grade patterns)

[$(pattern)] matches an atom that also matches the entire pattern starting from itself. Critical for context-aware matching.

# Aromatic carbon attached to a carbonyl
pat = Chem.MolFromSmarts('[$(c[C](=O))]')

# Aniline-type N (aromatic carbon-N-H)
pat = Chem.MolFromSmarts('[$([NX3;H2][c])]')

# Hindered amine (N with 2 sp3 neighbors)
pat = Chem.MolFromSmarts('[$([NX3]([CX4])([CX4])[CX4])]')

# H-bond donor (per Lipinski, exclude quaternary)
hbd = Chem.MolFromSmarts('[#7,#8;!H0;!$([NX3+])]')

# H-bond acceptor (per Lipinski, exclude nitro / aniline)
hba = Chem.MolFromSmarts('[$([#7,#8;!H0]);!$([NX3+]=O);!$(N(=O)~O)]')

Structural-Alert Filter Catalogs

When to Apply Each Filter

Critical: Capuzzi et al. (2017) showed that 8% of FDA-approved drugs match a PAINS pattern. PAINS is a flag for assay validation, not a killing filter.

PAINS Filter

Goal: Split a molecule list into PAINS-flagged and PAINS-clean sets using one or more PAINS catalog tiers.

Approach: Configure FilterCatalogParams with the requested catalog enums, build a FilterCatalog once, and for each molecule use GetFirstMatch to either bucket it as clean or record the matching pattern description.

from rdkit.Chem.FilterCatalog import FilterCatalog, FilterCatalogParams

def pains_filter(mols, catalogs=('PAINS_A',)):
    params = FilterCatalogParams()
    for cat in catalogs:
        params.AddCatalog(getattr(FilterCatalogParams.FilterCatalogs, cat))
    catalog = FilterCatalog(params)

    flagged = []
    clean = []
    for mol in mols:
        if mol is None:
            continue
        entry = catalog.GetFirstMatch(mol)
        if entry is None:
            clean.append(mol)
        else:
            flagged.append((mol, entry.GetDescription()))
    return clean, flagged

Available catalog names: PAINS_A, PAINS_B, PAINS_C, PAINS (all), BRENK, NIH, ZINC, ALL.

Reaction-Reactive Group Filter (custom)

For HTS triage, filter electrophilic warheads (acrylamide, chloroacetamide, etc.) unless designing covalent inhibitors.

Goal: Flag molecules containing electrophilic warheads or other reactive functional groups that would interfere with biochemical HTS.

Approach: Maintain a named SMARTS dictionary of reactive groups (acid halides, epoxides, Michael acceptors, etc.), then per molecule scan each pattern with HasSubstructMatch and return the first matching warhead name.

REACTIVE_SMARTS = {
    'acid_anhydride': '[CX3](=O)O[CX3](=O)',
    'acid_halide': '[CX3](=O)[F,Cl,Br,I]',
    'alpha_halo_carbonyl': '[CX3](=O)C([F,Cl,Br,I])',
    'aldehyde_reactive': '[CX3H1](=O)[#6;X4]',  # aliphatic aldehydes
    'epoxide': 'C1OC1',
    'aziridine': 'C1NC1',
    'isocyanate': '[NX2]=C=[OX1]',
    'isothiocyanate': '[NX2]=C=[SX1]',
    'beta_lactam': 'C1(=O)NCC1',
    'sulfonyl_halide': '[SX4](=O)(=O)[F,Cl,Br,I]',
    'Michael_acceptor': '[CX3]=[CX3][CX3]=O',
    'vinyl_sulfone': '[SX4](=O)(=O)C=C',
}

def reactive_filter(mol, exclude_warheads=True):
    if not exclude_warheads:
        return False
    for name, smarts in REACTIVE_SMARTS.items():
        if mol.HasSubstructMatch(Chem.MolFromSmarts(smarts)):
            return True, name
    return False, None

For covalent-inhibitor design, see chemoinformatics/covalent-design; these warheads are the desired chemistry, not noise to filter.

Library Filtering with Multiple Patterns

Goal: Reduce a molecule library to those that match all required SMARTS patterns and none of the excluded ones.

Approach: Start from the full molecule list, iteratively intersect with each include SMARTS using HasSubstructMatch, then subtract any molecule matching an exclude SMARTS.

def filter_library(mols, include=None, exclude=None):
    keep = list(mols)
    if include:
        for s in include:
            p = Chem.MolFromSmarts(s)
            keep = [m for m in keep if m and m.HasSubstructMatch(p)]
    if exclude:
        for s in exclude:
            p = Chem.MolFromSmarts(s)
            keep = [m for m in keep if m and not m.HasSubstructMatch(p)]
    return keep

Atom Map Indices in SMARTS

Atom maps [C:1] track atoms through transformations. Used in reactions (reaction-enumeration skill) but also for substructure-based extraction:

# Find amide N with attached aryl
pat = Chem.MolFromSmarts('[CX3:1](=O)[NX3:2][c:3]')
match = mol.GetSubstructMatch(pat)

amide_C, amide_N, aryl_C = match

Per-Tool Failure Modes

PAINS -- false positive on natural product

Trigger: Library contains natural products, polyphenols, flavonoids, quinones.

Mechanism: PAINS_A patterns target rhodanines, curcumins, polyhydroxylated polyphenols -- legitimate scaffolds in natural-product chemistry.

Symptom: Library hits flagged as PAINS but trace back to validated natural products with confirmed activity.

Fix: Use PAINS as a flag not a delete. Cross-check flagged compounds for orthogonal-assay confirmation (label-free e.g. SPR, ITC).

Aromaticity dialect mismatch

Trigger: SMARTS pattern with c (aromatic) for a heteroatom-rich ring; molecule parsed with different aromaticity model.

Mechanism: RDKit, OpenEye, ChemAxon differ on whether furan, thiazole, tropone, etc. are aromatic.

Symptom: Same pattern matches in one toolkit, not in another.

Fix: Re-canonicalize molecules within RDKit before applying SMARTS. Or use [#6]:[#6] instead of c:c (explicit element + bond type).

Tautomer-sensitive pattern miss

Trigger: SMARTS targets keto form C(=O) but molecule is enol C(O)=C.

Mechanism: Default canonical form differs by toolkit + standardization choice.

Symptom: Known matching molecule reports no match.

Fix: Use tautomer-aware match: enumerate tautomers and OR-match. Or canonicalize first via chemoinformatics/molecular-standardization. Or expand pattern with [$(C(=O)),$(C(O)=C)].

Stereochemistry ignored

Trigger: SMARTS without /\@ stereo markers applied to mol with explicit stereo.

Mechanism: SMARTS matching is stereo-agnostic by default.

Symptom: Wrong stereoisomer is matched as well as right one.

Fix: mol.GetSubstructMatches(pattern, useChirality=True) to require chirality match.

Ring closure / fused ring miss

Trigger: Pattern uses c1ccccc1 but target ring is fused (naphthalene, indole).

Mechanism: c1ccccc1 requires exactly 6 atoms in ring; not the fused-ring case.

Symptom: Naphthalene not matching benzene pattern.

Fix: Use ring-flexible pattern: [c]:[c]:[c]:[c]:[c]:[c] matches any aromatic 6-ring including fused. Or [c]1[c][c][c][c][c]1.

Recursive SMARTS performance

Trigger: Deeply nested recursive SMARTS over a large library.

Mechanism: Each [$()] re-evaluates the inner pattern for every candidate atom.

Symptom: Search 10x-100x slower than expected.

Fix: Flatten recursion where possible; pre-filter with simpler pattern, then re-test with the recursive one.

Common Errors

References

• Baell & Holloway, J. Med. Chem. 53:2719 (2010) -- original PAINS filter.
• Capuzzi et al., J. Chem. Inf. Model. 57:417 (2017) -- PAINS reality check (FDA drug overlap).
• Brenk et al., ChemMedChem 3:435 (2008) -- structural alerts (BRENK filter).
• Walters & Murcko, Adv. Drug Deliv. Rev. 54:255 (2002) -- REOS filter framework.
• Bruns & Watson, J. Med. Chem. 55:9763 (2012) -- Eli Lilly medchem rules.
• Daylight SMARTS theory documentation -- complete grammar reference.

Related Skills

• chemoinformatics/molecular-io - Parse molecules before searching
• chemoinformatics/molecular-standardization - Canonicalize tautomers before SMARTS
• chemoinformatics/similarity-searching - Fingerprint-based fuzzy matching
• chemoinformatics/scaffold-analysis - Scaffold-based pattern derivation
• chemoinformatics/reaction-enumeration - SMARTS for chemical transformations
• chemoinformatics/admet-prediction - PAINS as ADMET filter
• chemoinformatics/covalent-design - Warhead chemistry