Background/Objectives: Rules for predicting blood–brain barrier (BBB) permeability, including the CNS multiparameter optimization (CNS MPO) score, Lipinski’s Rule of Five, and Veber’s rules, were developed using relatively limited datasets and have not been systematically re-evaluated using modern large-scale experimental databases. Using the B3DB database, which contains 1058 experimentally measured logBB values, we derived quantitative, data-driven structural thresholds for BBB permeability and benchmarked them against established heuristic rules. Methods: Six key physicochemical properties were calculated for all compounds, and optimal classification thresholds were identified through exhaustive optimizations. Decision trees and scaffold analyses were used to generate interpretable medicinal chemistry guidelines. Results: The topological polar surface area (TPSA) emerged as the strongest single predictor of BBB permeability, with an optimal threshold of 66.8 Å2 (AUC = 0.731, 95% CI: 0.689–0.771). This threshold outperformed the approximated CNS MPO ≥ 4 (AUC = 0.625), Lipinski’s Rule of Five (AUC = 0.546), and Veber rules (AUC = 0.566). A simple two-parameter rule combining TPSA < 67 Å2 and H-bond donors ≤ 1 achieved 96.6% precision for BBB-permeable compounds while maintaining an AUC of 0.720. Decision tree analysis further confirmed TPSA as the dominant determinant of BBB permeability, whereas scaffold analysis identified the molecular frameworks associated with highly permeable and impermeable compounds. External validation provided preliminary support for the improved specificity of the proposed rule, compared with existing approaches. Conclusions: These findings suggest that the commonly applied TPSA threshold of 90 Å2 may be overly lenient. A data-driven threshold of approximately 67 Å2 substantially improved the discrimination of BBB permeability across the entire dataset. Compounds with TPSA values between 67 and 90 Å2 should be assessed on a case-by-case basis, considering the ionization state and active transport potential, rather than being automatically classified as BBB-permeable. These experimentally grounded rules offer a practical framework for the early-stage design of CNS leads.
S. Tiwari, Katarzyna Mądra-Gackowska, Marcin Gackowski et al.· Pharmaceutics· 0 citations
Background/Objectives: Predicting blood–brain barrier (BBB) permeability remains a major challenge in central nervous system (CNS) drug discovery. Three-dimensional (3D) conformer-derived molecular descriptors are often proposed as improvements over conventional two-dimensional (2D) topological representations; however, their incremental predictive value remains unclear. Methods: Here, we systematically evaluated six molecular feature sets comprising curated 2D Mordred descriptors, Morgan/ECFP4 fingerprints, 3D conformer-derived descriptors, and their combinations, using LightGBM regression on the B3DB benchmark dataset (1054 compounds with experimental logBB values). The model performance was assessed using 30 independent random splits and 30 Murcko scaffold-based splits. Results: The 2D descriptor model achieved mean test R2 values of 0.567 ± 0.060 and 0.418 ± 0.085 under random and scaffold splitting, respectively, whereas the 3D descriptor model performed substantially worse (0.430 ± 0.066 and 0.298 ± 0.097, respectively). Incorporating 3D descriptors into the 2D feature set yielded only a marginal improvement under random splitting (+0.008 R2; p = 0.039), which disappeared during scaffold-based validation (p = 0.599). Similarly, adding 3D descriptors to the combined 2D + fingerprint representation did not yield any significant benefits. Fingerprints alone exhibited pronounced scaffold fragility, with the mean R2 decreasing from 0.441 to 0.312 between the random and scaffold evaluations. SHAP analysis identified TopoPSA (NO) as the dominant predictor (mean |SHAP| = 0.332), highlighting the central role of desolvation in BBB permeation. Applicability domain analysis showed that 94.8% of the test compounds fell within the structural coverage of the training set. Conclusions: Overall, on the B3DB benchmark and under the descriptor implementation evaluated here, 3D conformer descriptors provided limited incremental value and no scaffold-robust advantage over well-curated 2D molecular representations for BBB permeability predictions.
S. Tiwari, Katarzyna Mądra-Gackowska, Marcin Gackowski et al.· Pharmaceutics· 0 citations
The growing demand for efficient energy conversion and advanced optoelectronic devices has driven extensive research into multifunctional materials with tunable physical properties. In this work, we present a comprehensive first-principles investigation of cubic perovskites ZnAgX3 (X = F, Cl, Br) using density functional theory as implemented in the Quantum ESPRESSO package. The structural, electronic, optical, mechanical, and thermoelectric properties are systematically explored employing both GGA-PBE and HSE06 exchange–correlation functionals. Structural optimization confirms that all compounds crystallize in the stable cubic Pm3̅m phase, with thermodynamic stability evidenced by negative formation energies ranging from −2.109 to −1.176 eV, Goldschmidt tolerance factors of 0.891–0.848, and dynamical stability verified by the absence of imaginary phonon modes. Electronic band structure calculations reveal indirect semiconducting behavior with tunable band gaps that decrease from ZnAgF3 to ZnAgBr3. The HSE06 functional predicts band gaps of 2.857 eV, 2.503 eV, and 1.673 eV for ZnAgF3, ZnAgCl3, and ZnAgBr3, respectively. Optical analysis indicates a strong halogen-dependent response in dielectric function, refractive index, absorption, and optical conductivity, with ZnAgBr3 exhibiting the highest static dielectric constant (3.86) and absorption coefficient (∼1.70 × 106 cm−1) due to its narrower band gap. Mechanical studies have verified that all the compounds meet the Born stability requirements and that they are ductile. The calculated B/G ratios (2.88–3.07) further confirm the ductile nature of all investigated compounds. Thermoelectric analysis demonstrates a significant enhancement in performance with heavier halide substitution, with ZnAgBr3 achieving a high figure of merit (ZT = 0.98 at 300 K and ZT ≈1.96 at 1000 K), highlighting its potential for high-temperature thermoelectric applications. Overall, halogen substitution is shown to be an effective strategy for tailoring the bonding characteristics, electronic structure, and transport properties, positioning ZnAgBr3 as a promising candidate for multifunctional optoelectronic and thermoelectric devices.
Łukasz Szeleszczuk, Katarzyna Mądra-Gackowska, Marcin Gackowski· Physica Scripta· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
