Dataset and Analysis Code for: Cheminformatics Analysis of Plasticizer Chemical Space Reveals Gaps in Nigerian Environmental Monitoring

Introduction:

This repository contains the complete dataset, analysis code, and supplementary tables associated with the manuscript:

"Cheminformatics Analysis of Plasticizer Chemical Space Reveals Gaps in Nigerian Environmental Monitoring"

Contents:

1. Phthalate_Dataset_final.xlsx A curated virtual library of 40 commercially relevant plasticizers spanning four structural categories: core phthalate diesters (n=9), extended phthalate homologues (n=9), monoester metabolites and degradation products (n=10), and emerging alternative plasticizers (n=12). For each compound the file contains: compound name, abbreviation, CAS number, molecular formula, structural category, eleven RDKit-calculated physicochemical descriptors (MW, MolLogP, TPSA, HBD, HBA, RotBonds, RingCount, AromaticRings, HeavyAtoms, FractionCSP3, MolMR), Nigerian reporting status (Y/NR), hierarchical cluster assignment, PCA scores (PC1, PC2), and canonical SMILES string.

2. Tables_1_and_S1.xlsx Two formatted tables:

• Table 1: PCA component loadings for the nine retained physicochemical descriptors
• Table S1: Nigerian peer-reviewed literature sources used to classify compounds as Reported (Y) or Not Reported (NR), including reference, year, Nigerian location, environmental matrix, compounds detected, and DOI

3. phthalate_cheminformatics_clean.ipynb A fully reproducible Jupyter Notebook (Python 3.11) containing the complete analysis workflow: SMILES validation, descriptor calculation, correlation-based feature selection, PCA, hierarchical Ward linkage clustering, statistical comparisons (Mann-Whitney U, Cohen's d, Fisher's exact test), and generation of all manuscript figures. Run sequentially using Kernel → Restart & Run All.

Software requirements:

• Python 3.11
• RDKit 2025.03.6
• scikit-learn 1.8.0
• pandas 3.0.3
• scipy
• matplotlib
• seaborn

Keywords: phthalates, plasticizers, cheminformatics, principal component analysis, hierarchical clustering, Nigeria, environmental monitoring, chemical space, RDKit, molecular descriptors

Keywords field (paste these individually into Zenodo's keyword boxes): phthalates — plasticizers — cheminformatics — PCA — Nigeria — environmental monitoring — chemical space — RDKit — molecular descriptors — Ward clustering

License: Creative Commons Attribution 4.0 International (CC BY 4.0)

Related publication: Add the journal DOI here once the paper is accepted.