Nicola Gambacorta

QT interval prolongation is associated with an increased risk of torsades de pointes (TdP), a potentially life-threatening ventricular arrhythmia, making assessment of cardiac repolarization a critical component of drug development. This work demonstrates how integrated in silico methodologies can be used to predict and quantitatively characterize QT liability across the drug discovery and development continuum. A proof-of concept translational workflow combining molecular structure analysis, in vitro pharmacology, animal data, and clinical exposure-response modeling was evaluated using the retrospective case examples verapamil and dofetilide. Molecular structure- and pharmacology-based assessments predicted hERG channel interaction and potential QT risk for both compounds. While initial in silico analyses suggested a high QT liability for verapamil and dofetilide, mechanistic electrophysiological modeling differentiated their effects. For verapamil, concurrent calcium channel inhibition offset hERG-mediated effects, resulting in predicted reductions in action potential duration and no clinically relevant QT prolongation. In contrast, dofetilide produced concentration-dependent increases in action potential duration, consistent with its known proarrhythmic potential. Translation of nonclinical findings to humans accurately predicted the absence of clinically meaningful QT effects for verapamil and identified QT prolongation exceeding the regulatory 10 ms threshold for dofetilide at therapeutic concentrations. Drug exposure and pharmacokinetic properties could be anticipated based solely on molecular structure. This conceptual workflow demonstrates how integrated quantitative in silico approaches can support earlier identification, characterization, and communication of QT risk, enabling informed decision-making and improved development efficiency from bench to bedside.

Toxicity prediction in small molecules represents a fundamental challenge in drug development and chemical safety assessment. Traditional approaches heavily rely on predefined molecular descriptors or fingerprints, potentially limiting the ability to capture complex and nonlinear structure–activity relationships. Here, we present a descriptor-free, language-inspired framework that can be applied to different toxicity prediction tasks within a unified architecture. The model proposed combines a multiscale Convolutional Neural Network (CNN) layer to capture chemical patterns at different scales and a Gated Recurrent Unit (GRU) layer to capture the sequential nature of these patterns. This architecture also exploits an attention mechanism that computes attention weights across the sequence, enabling the model to focus on the most relevant molecular substructures for toxicity prediction. Toxic and nontoxic chemicals, represented by canonical SMILES, are investigated as the words of two languages which have to be discriminated; using eight different end points, the model provided an accurate description of toxicity patterns, with an average Area under the ROC curve (AUC) of 0.83 (min: 0.70, max: 0.94) under repeated cross-validation. The models were trained on relatively small data sets (∼1000 samples) and often strongly imbalanced, two important challenges that highlight the opportunities for future improvement; moreover, the proposed attention-based framework offers a representation of the molecular regions influencing model predictions, providing a basis for future investigations into toxicity-related structural patterns and potentially supporting hypothesis generation in drug design or drug repurposing applications.

Alzheimer’s disease (AD) is a multifactorial neurodegenerative disorder involving a complex interplay of interconnected pharmacological targets, a fe

Endocrine disruption remains a major concern in predictive toxicology, demanding accurate and interpretable models to assess molecular interactions with hormonal pathways. Here, we present a descriptor-free bidirectional long short-term memory (BiLSTM) framework designed to predict the toxicity of chemicals toward androgen (AR) and estrogen receptors (ER), two of the most complex and biologically relevant end points in toxicology. The model operates directly on SMILES strings, which are tokenized and converted into one-hot encoded sequences, enabling the automatic extraction of chemically meaningful representations without reliance on handcrafted molecular descriptors or fingerprints. To enhance interpretability, we introduce a novel explainable artificial intelligence (XAI) approach that aggregates character-level attribution scores into color-coded substructures, revealing features that drive or reduce toxicity and offering mechanistic insight into receptor-mediated effects. The models were trained on publicly available, high-quality data sets comprising 1664 and 1529 chemicals with experimental binary labels for AR and ER, respectively. Employing cross-validation analyses, based on 20% randomly stratified resampling iterated 10 times, the proposed workflow returned accuracy equal to 0.75 ± 0.08 and 0.81 ± 0.05, sensitivity equal to 0.66 ± 0.36 and 0.69 ± 0.17, and specificity equal to 0.76 ± 0.14 and 0.82 ± 0.06 for AR and ER end points, respectively. Our descriptor-free models ensure highly transparent results with a substructure level interpretability. These findings demonstrate the potential of deep learning directly on molecular textual representations to advance predictive toxicology and to support mechanistic understanding in chemical risk assessment.

Background: Lafora disease (LD) is an ultra-rare and fatal neurodegenerative disorder with limited therapeutic options. Current treatments primarily address symptoms, with modest efficacy in halting disease progression, thus highlighting the urgent need for novel therapeutic approaches. Gene therapy, antisense oligonucleotides, and recombinant enzymes have recently been, and still are, under investigation. Drug repurposing may offer a promising approach to identify new, possibly effective, therapies. Methods: This study aims to investigate the conditions for repurposing empagliflozin, an SGLT2 (sodium/glucose cotransporter-2) inhibitor, as a potential treatment for LD and to establish a clinical protocol. Clinical phase: This 12-month prospective observational study will assess the safety and clinical efficacy of empagliflozin in two patients with early to intermediate LD stage. The primary endpoints will include changes in the severity of epilepsy and cognitive function, while the secondary endpoints will assess motor function, global function, and autonomy. Multiple clinical and instrumental evaluations (including MRI and PET with ¹⁸F-fluorodeoxyglucose) will be performed before and during treatment. Safety monitoring will include regular clinical assessments and reports of adverse events. Preclinical phase: In silico studies (using both molecular docking calculations and reverse ligand-based screening) and in vitro cell-based assays will allow us to investigate the effects of empagliflozin (and other gliflozins) on some key targets likely implicated in LD pathogenesis, such as GLUT1, GLUT3, glycogen synthase (hGYS), and glycogen phosphorylase (GP), as suggested in the literature and digital platforms for in silico target fishing. Results: The expected outcome of this study is twofold, i.e., (i) assessing the safety and tolerability of empagliflozin in LD patients and (ii) gathering preliminary data on its potential efficacy in improving clinical and neurologic features. Additionally, the in silico and in vitro studies may provide new insights into the mechanisms through which empagliflozin may exert its therapeutic effects in LD. Conclusion: The findings of this study are expected to provide evidence in support of the repurposing of empagliflozin for the treatment of LD.

The advent of powerful machine learning algorithms as well as the availability of high volume of pharmacological data has given new fuel to QSAR, opening new unprecedented options for deriving highly predictive models for assisting the rationale design of new bioactive compounds, for screening and prioritizing large molecular libraries, and for repurposing new drugs toward new clinical uses. Here, we present PoseidonQ (an acronym for Personal Optimization Software for Efficient Implementation and Derivation of Online QSAR), a user-friendly software solution designed to simplify the derivation of the QSAR model for drug design and discovery. PoseidonQ incorporates 22 machine learning algorithms, 17 types of molecular fingerprints, and 208 RDKit molecular descriptors and enables the quick derivation of both regression and classification models along with a calculated and easily interpretable applicability domain. Importantly, the platform is automatically linked to the latest version of the ChEMBL database, thus providing streamlined access to large amounts of curated bioactivity data. Importantly, the user is also given the option of gathering high-quality experimental data based on customizable filtering settings. Noteworthy, PoseidonQ facilitates the deployment of trained QSAR models as web-based applications through seamless integration with Streamlit Cloud and GitHub, empowering users to share, refine, and integrate models effortlessly. Interestingly, the translation of QSAR models into web-based applications makes them free accessible, portable, and ready for screening large volumes of new data without limits. By unifying data preparation, model generation, and deployment into an intuitive workflow, PoseidonQ makes advanced QSAR modeling for drug design and discovery accessible to a wide audience of researchers irrespective of their skill levels. PoseidonQ bridges the gap between complex machine learning techniques and practical drug discovery applications, enhancing the efficiency, collaboration, and adoption of QSAR approaches in modern drug discovery programs. PoseidonQ is available for Windows and Linux (ubuntu 22.04 distro) operating systems and can be downloaded for free at https://github.com/Muzatheking12/PoseidonQ.

Chemical space networks (CSNs) are a new effective strategy for detecting latent chemical patterns irrespective of defined coordinate systems based on molecular descriptors and fingerprints. CSNs can be a new powerful option as a new approach method and increase the capacity of assessing potential adverse impacts of chemicals on human health. Here, CSNs are shown to effectively characterize the toxicity of chemicals toward several human health end points, namely chromosomal aberrations, mutagenicity, carcinogenicity, developmental toxicity, skin irritation, estrogenicity, androgenicity, and hepatoxicity. In this work, we report how the content from CSNs structure can be embedded through graph neural networks into a metric space, which, for eight different toxicological human health end points, allows better discrimination of toxic and nontoxic chemicals. In fact, using embeddings returns, on average, an increase in predictive performances. In fact, embedding employment enhances the learning, leading to an increment of the classification performance of +12% in terms of the area under the ROC curve. Moreover, through a dedicated eXplainable Artificial Intelligence framework, a straight interpretation of results is provided through the detection of putative structural alerts related to a given toxicity. Hence, the proposed approach represents a step forward in the area of alternative methods and could lead to breakthrough innovations in the design of safer chemicals and drugs.

Generative models have revolutionized de novo drug design, allowing to produce molecules on-demand with desired physicochemical and pharmacological properties. String based molecular representations, such as SMILES (Simplified Molecular Input Line Entry System) and SELFIES (Self-Referencing Embedded Strings), have played a pivotal role in the success of generative approaches, thanks to their capacity to encode atom- and bond- information and ease-of-generation. However, such ‘atom-level’ string representations could have certain limitations, in terms of capturing information on chirality, and synthetic accessibility of the corresponding designs. In this paper, we present fragSMILES, a novel fragment-based molecular representation in the form of string. fragSMILES encode fragments in a ‘chemically-meaningful’ way via a novel graph-reduction approach, allowing to obtain an efficient, interpretable, and expressive molecular representation, which also avoids fragment redundancy. fragSMILES contributes to the field of fragment-based representation, by reporting fragments and their ‘breaking’ bonds independently. Moreover, fragSMILES also embeds information of molecular chirality, thereby overcoming known limitations of existing string notations. When compared with SMILES, SELFIES and t-SMILES for de novo design, the fragSMILES notation showed its promise in generating molecules with desirable biochemical and scaffolds properties.

The 2-oxoglutarate carrier (OGC), pivotal in cellular metabolism, facilitates the exchange of key metabolites between mitochondria and cytosol. This study explores the influence of NADPH on OGC transport activity using proteoliposomes. Experimental data revealed the ability of NADPH to modulate the OGC activity, with a significant increase of 60% at 0.010 mM. Kinetic analysis showed increased Vmax and a reduction in Km for 2-oxoglutarate, suggesting a direct regulatory role. Molecular docking pointed to a specific interaction between NADPH and cytosolic loops of OGC, involving key residues such as K206 and K122. This modulation was unique in mammalian OGC, as no similar effect was observed in a plant OGC structurally/functionally related mitochondrial carrier. These findings propose OGC as a responsive sensor for the mitochondrial redox state, coordinating with the malate/aspartate and isocitrate/oxoglutarate shuttles to maintain redox balance. The results underscore the potential role of OGC in redox homeostasis and its broader implications in cellular metabolism and oxidative stress responses.