SuperAPI AI Directory

scientific-cheminformatics

General & other

View repository

Updated 10-15 21:04

About this skill

Scientific Cheminformatics Analysis

Molecular analysis pipeline skills using RDKit. Provides a drug discovery chemoinformatics workflow from SMILES → molecular descriptors → SAR analysis → toxicity prediction.

When to Use

  • When you want to calculate the physicochemical properties of a compound
  • When you want to calculate molecular descriptors from a SMILES string
  • When you want to evaluate structural similarity between compounds
  • When you want to analyze structure-activity relationships (SAR)
  • When you want to detect structural alerts (toxicophores)
  • When you want to evaluate Lipinski Rule of 5 / drug-likeness

Quick Start

Standard pipeline

1. SMILES → molecular object conversion

from rdkit import Chem
from rdkit.Chem import Descriptors, AllChem, QED, Lipinski
from rdkit.Chem.Scaffolds import MurckoScaffold
import pandas as pd
import numpy as np

def smiles_to_mol(smiles):
    """SMILES 文字列から RDKit 分子オブジェクトを生成する。"""
    mol = Chem.MolFromSmiles(smiles)
    if mol is None:
        raise ValueError(f"Invalid SMILES: {smiles}")
    return mol

2. Batch calculation of molecular descriptors

def calculate_descriptors(smiles_list, names=None):
    """
    SMILES リストから主要な分子記述子を一括計算する。
    返値: DataFrame
    """
    records = []
    for i, smi in enumerate(smiles_list):
        mol = Chem.MolFromSmiles(smi)
        if mol is None:
            continue

        record = {
            "Name": names[i] if names else f"Mol_{i}",
            "SMILES": smi,
            "MW": Descriptors.MolWt(mol),
            "LogP": Descriptors.MolLogP(mol),
            "TPSA": Descriptors.TPSA(mol),
            "HBA": Descriptors.NumHAcceptors(mol),
            "HBD": Descriptors.NumHDonors(mol),
            "RotBonds": Descriptors.NumRotatableBonds(mol),
            "AromaticRings": Descriptors.NumAromaticRings(mol),
            "HeavyAtoms": mol.GetNumHeavyAtoms(),
            "QED": QED.qed(mol),
            "Fraction_CSP3": Descriptors.FractionCSP3(mol),
        }
        records.append(record)

    return pd.DataFrame(records)

3. Morgan fingerprint & Tanimoto similarity

from rdkit import DataStructs

def compute_fingerprints(smiles_list, radius=2, nBits=2048):
    """Morgan フィンガープリントを生成する。"""
    fps = []
    for smi in smiles_list:
        mol = Chem.MolFromSmiles(smi)
        if mol:
            fp = AllChem.GetMorganFingerprintAsBitVect(mol, radius, nBits=nBits)
            fps.append(fp)
    return fps

def tanimoto_similarity_matrix(fps, names=None):
    """Tanimoto 類似度行列を算出する。"""
    n = len(fps)
    sim_matrix = np.zeros((n, n))
    for i in range(n):
        for j in range(n):
            sim_matrix[i, j] = DataStructs.TanimotoSimilarity(fps[i], fps[j])

    if names is None:
        names = [f"Mol_{i}" for i in range(n)]

    sim_df = pd.DataFrame(sim_matrix, index=names, columns=names)
    sim_df.to_csv("results/tanimoto_similarity.csv")
    return sim_df

4. Lipinski Rule of 5 evaluation

def lipinski_evaluation(desc_df):
    """Lipinski Rule of 5 の準拠チェック。"""
    desc_df = desc_df.copy()
    desc_df["Lipinski_MW"] = desc_df["MW"] <= 500
    desc_df["Lipinski_LogP"] = desc_df["LogP"] <= 5
    desc_df["Lipinski_HBA"] = desc_df["HBA"] <= 10
    desc_df["Lipinski_HBD"] = desc_df["HBD"] <= 5
    desc_df["Lipinski_Violations"] = 4 - (
        desc_df["Lipinski_MW"].astype(int) +
        desc_df["Lipinski_LogP"].astype(int) +
        desc_df["Lipinski_HBA"].astype(int) +
        desc_df["Lipinski_HBD"].astype(int)
    )
    desc_df["Lipinski_Pass"] = desc_df["Lipinski_Violations"] <= 1
    return desc_df

5. Structural alert (toxicophore) detection (Exp-05)

STRUCTURAL_ALERTS = {
    "Nitro": "[N+](=O)[O-]",
    "Epoxide": "C1OC1",
    "Aldehyde": "[CH]=O",
    "Michael_Acceptor": "C=CC(=O)",
    "Acyl_Halide": "C(=O)[F,Cl,Br,I]",
    "Aniline": "c1ccccc1N",
    "Hydrazine": "NN",
    "Sulfonate": "S(=O)(=O)[O-]",
}

def detect_structural_alerts(smiles_list, names=None, alerts=None):
    """SMARTS パターンによる構造アラートの検出。"""
    if alerts is None:
        alerts = STRUCTURAL_ALERTS

    results = []
    for i, smi in enumerate(smiles_list):
        mol = Chem.MolFromSmiles(smi)
        if mol is None:
            continue

        name = names[i] if names else f"Mol_{i}"
        for alert_name, smarts in alerts.items():
            pattern = Chem.MolFromSmarts(smarts)
            if mol.HasSubstructMatch(pattern):
                results.append({"Name": name, "SMILES": smi,
                               "Alert": alert_name, "SMARTS": smarts})

    return pd.DataFrame(results)

6. Murcko scaffold analysis

def scaffold_analysis(smiles_list, names=None):
    """Murcko スキャフォールドの抽出と分類。"""
    scaffolds = []
    for i, smi in enumerate(smiles_list):
        mol = Chem.MolFromSmiles(smi)
        if mol:
            core = MurckoScaffold.GetScaffoldForMol(mol)
            scaffolds.append({
                "Name": names[i] if names else f"Mol_{i}",
                "SMILES": smi,
                "Scaffold": Chem.MolToSmiles(core),
            })
    return pd.DataFrame(scaffolds)

References

Output Files

File Format
results/molecular_properties.csv CSV
results/tanimoto_similarity.csv CSV
results/structural_alerts.csv CSV
figures/chemical_space_pca.png PNG
figures/similarity_heatmap.png PNG

Available tools

ToolUniverse External tools available via SMCP.

Category Main tool Purpose
PubChem PubChem_get_CID_by_compound_name Compound name → CID conversion
PubChem PubChem_get_compound_properties_by_CID Compound property retrieval
PubChem PubChem_search_compounds_by_similarity Similar compound search
ChEMBL ChEMBL_search_molecules Molecular search
ChEMBL ChEMBL_get_molecule Molecular information retrieval
ZINC ZINC_search_by_smiles SMILES-based search

Reference experiments

  • Exp-02: EGFR inhibitor SAR analysis (descriptors, Tanimoto, MCS, Scaffold)
  • Exp-05: Toxicity prediction (structural alerts, Morgan FP classification model)