Achieving optimal target activity while maintaining synthetic accessibility and drug-likeness represents a major challenge in computational drug discovery. Existing de novo generative models often yield chemically invalid or synthetically intractable structures and struggle to optimize multiple objectives simultaneously. Here, we introduce ALCHIMIA, an interpretable hybrid framework combining reinforcement learning (RL) and a genetic algorithm (GA), built based on a vocabulary of 33 medicinal chemistry-inspired molecular transformations. The RL component trains a policy network to prioritize transformation sequences that improve synthetic accessibility (SA) and the quantitative estimate of drug-likeness (QED) scores, embedding these constraints directly into molecular generation. The GA component applies the learned policy as a mutational operator within population-based optimization guided by molecular docking, enabling the exploration of diverse chemical lineages while converging toward high-affinity ligands. ALCHIMIA was applied to two different pharmacologically relevant targets: human Cannabinoid Receptor 2 (CB2R) and human Sigma nonopioid intracellular Receptor 1 (S1R). We considered three different scenarios: (i) unconstrained hit identification; (ii) scaffold-constrained lead optimization; and (iii) design of dual modulators. The framework generated chemically valid molecules with QED and SA scores comparable to or better than those obtained with random baselines and selected de novo design methods. By codifying typical medicinal chemistry actions as learnable transformations and coupling multiobjective optimization with GA-based diversity maintenance, ALCHIMIA, freely available as a GitHub repository (https://github.com/alberdom88/ALCHIMIA), provides a practical, interpretable, and scalable framework for molecular de novo design.
This study reports the discovery and structural characterization of a novel polymorph, designated as Form III, of Alclometasone dipropionate, a corticosteroid commonly used in the treatment of inflammatory dermatoses. Form III was obtained by modifying the crystallization conditions reported in prior art and was thoroughly characterized using Powder X-ray Diffraction (PXRD), Fourier Transform Infrared (FT-IR) spectroscopy, melting-point determination, Differential Scanning Calorimetry (DSC), Thermogravimetric Analysis (TGA), including its first derivative (DTG), optical microscopy, and Scanning Electron Microscopy (SEM). In parallel, pure Form II, previously observed only in mixtures with Form I, was successfully isolated and characterized using the same analytical techniques. Both forms were compared in terms of structural, thermal, and morphological properties. PXRD analysis revealed that Form III crystallizes in a triclinic system; FT-IR spectroscopy revealed unique vibrational signatures, and microscopy showed rod-like crystal morphology. The discovery of Form III expands the current understanding of the solid-state landscape of Alclometasone dipropionate and opens opportunities for the identification of new industrial purification methods for the compound.
11 October 2026 in Journal of Cellular and Molecular Medicine
Mucopolysaccharidosis Type I (MPS I) is a rare autosomal recessive lysosomal storage disorder caused by mutations in α‐ʟ‐iduronidase (IDUA) enzyme, a glycosylated hydrolase responsible for glycosaminoglycan (GAG) degradation. Pathogenic variants are associated with a broad clinical spectrum, from severe Hurler syndrome to attenuated Scheie phenotype. The missense mutations L490P and P533R cause attenuated‐intermediate MPS I, retaining residual enzyme activity. To investigate the structural mechanisms underlying these phenotypes, we performed all‐atom molecular dynamics simulations –18 trajectories of 1.6 μs each– comparing wild type IDUA with L490P and P533R mutants at pH 7.5 and 4.5, mimicking the endoplasmic reticulum (ER) and lysosomal microenvironments, respectively. Backbone root‐mean‐square deviation and fluctuation analyses indicated that the overall protein fold was maintained across conditions; however, we identified a pH‐dependent restricted conformational flexibility of the N372‐linked glycan in both mutants. Cluster analysis performed on the obtained trajectories showed that both mutations affect the conformation of this glycan. Hydrogen bond analysis further revealed that P533R induces aberrant, persistent glycan–protein interactions, providing a possible structural explanation for reduced IDUA activity and, therefore, GAG processing. Our findings, supported by in vitro data, propose glycan conformational restriction as a possible new molecular mechanism that underpins enzymatic dysfunction in attenuated MPS I, opening new avenues for structure‐informed therapeutic strategies.
Phosphoprotein Enriched in Diabetes/Astrocytes 15 (PED/PEA-15) is a 15 kDa adaptor protein whose overexpression drives tumorigenesis and type II diabetes through dysregulated protein-protein interactions (PPIs) mediated by its Death Effector Domain (DED). Despite its therapeutic relevance, no potent small-molecule inhibitors targeting the DED have yet been identified. Here, we employed a fragment-based drug discovery approach combining 19F and 1H NMR screening of a 118-compound fluorinated library to identify F2 a fragment that binds the PED/PEA-15 DED with an estimated dissociation constant of 150 μM. NMR chemical shift perturbation analysis mapped the binding site to the α2-α3 loop, α3, and α5 helices. To better characterise the protein, Small-angle X-ray scattering of the apo protein was employed, yielding a structural model in which the C-terminal tail is consistent with a conformation that is not maximally extended. Induced fit docking and molecular dynamics simulations of the PED/PEA-15-F2 complex supported a plausible binding pose and suggested a ligand-induced destabilisation of the α3 helix propagating to the α6 helix. Correlation network analysis further suggests a hypothesis in which F2 rewires the dynamic architecture of the DED and alters allosteric communication between residues of the DED and the C-terminal tail. The identification of a ligand-accessible site on the DED supports the druggability of PED/PEA-15 and provides a refined structural framework for fragment optimisation and for testing whether engagement of this site can be translated into modulation of PED/PEA-15 PPIs with therapeutic potential in cancer and type II diabetes.