ProQSAR: A modular and reproducible framework for small-data QSAR modeling with fit-and-use models

Quantitative structure-activity relationship (QSAR) models are central to computer-aided drug discovery and predictive toxicology, but practical adoption is often impeded by ad-hoc tooling, inconsistent validation protocols, and poor reproducibility.

We introduce ProQSAR, a modular, reproducible workbench that formalizes end-to-end QSAR development while permitting independent use of each component.

ProQSAR composes interchangeable modules for standardization, feature generation, splitting (including scaffold- and cluster-aware splits), preprocessing, outlier handling, scaling, feature selection, model training and tuning, statistical comparison, conformal calibration, and applicability-domain assessment. The pipeline can run end-to-end to produce versioned artifact bundles (serialized models) and analyst-oriented reports suitable for deployment and audit.

On representative MoleculeNet benchmarks evaluated under Bemis-Murcko scaffold split, ProQSAR attains state-of-the-art descriptor-based performance: the lowest mean RMSE across the regression suite (ESOL, FreeSolv, Lipophilicity; mean RMSE $0.658\pm 0.11$ ), including a substantial improvement on FreeSolv (RMSE $0.494$ vs. $0.731$ for a leading graph method). On quantum mechanical benchmarks, ProQSAR demonstrated superior performance on the single-task dataset QM7 and maintained competitive results on the multi-task QM8 dataset. For classification, ProQSAR achieves the top ROC-AUC on ClinTox (91.4%) while remaining competitive across other benchmark (overall classification average $70.4\pm 11.6$ ). Crucially, all predictions are accompanied by cross-conformal prediction and explicit applicability-domain flags that identify out-of-distribution entries, enabling calibrated and decision support.

ProQSAR is released on PyPI, Conda, and Docker Hub; all releases embed full provenance (parameters, package versions, checksums) to ensure reproducibility.

ProQSAR (i) enforces best-practice, group-aware validation together with formal statistical comparisons across models, (ii) integrates calibrated uncertainty quantification (cross-conformal prediction) and applicability-domain diagnostics for interpretable, risk-aware predictions, and (iii) exposes both a composable developer API and a one-click pipeline that generates deployment-ready artifacts and human-readable reports, demonstrated on representative benchmarks.