arXiv subjects

Pawel Dabrowski-Tumanski

Publications and source records attributed to Pawel Dabrowski-Tumanski.

4 recordsLinked to original sources

arXiv2026

The Geometry of Activity Cliffs: Representation Dependence and Multi-Scale Characterization of Activity Landscapes

Activity cliffs, structurally similar compounds with large potency differences, are widely treated as intrinsic features of chemical datasets. We argue that apart from target biology, much of our cliff understanding is a consequence of the geometry induced by the chosen molecular representation, not a property of a molecule pair itself. We designed a six-step pipeline to systematically test this hypothesis. The pipeline consists of: assessing pairwise distance geometry, cliff enrichment, activity gradient distribution, persistent homology of the cliff subspace, predictive benchmarking for a chosen pair of an embedding and a metric, and eventually, analysis of the matched molecular pairs and stereoisomers. We applied the pipeline to fifteen configurations of embeddings and metrics to build a benchmark across three distinctive datasets known of activity cliffs challenges. No representation excels on all criteria: Morgan Tanimoto provides the strongest cliff enrichment and cross-scaffold generalization; MolFormer cosine provides the only meaningful stereochemical sensitivity; MACCS and RDKit Dice fingerprints are most sensitive to matched-molecular-pair transformations; ChemBERTa fails uniformly due to embedding collapse. These findings are not a ranking. They reflect the fact that different representations encode different aspects of molecular recognition, and that choosing one implicitly defines what an activity cliff actually is.

q-bio.QM↗

arXiv2026

An explainable hypothesis-driven approach to Drug-Induced Liver Injury with HADES

Drug-induced liver injury (DILI) remains a leading cause of late-stage clinical trial attrition. However, existing computational predictors primarily rely on binary classification, a framing that limits generalization and yields no mechanistic insight to guide translational decisions. We argue that DILI prediction is better posed as an explainable hypothesis-generation problem. To support this shift, we introduce the DILER Benchmark, a dataset that extends beyond binary labels by augmenting a curated set of molecules with mechanistic hepatotoxicity hypotheses derived from biomedical literature. We further present HADES, an agentic system designed to generate transparent and auditable reasoning traces. By combining molecular-level predictions, metabolite decomposition, structural understanding, and toxicity pathway evidence, HADES mechanistically assesses DILI risk. Evaluated on the DILER Benchmark, HADES outperforms existing models in binary classification, achieving a ROC-AUC of 0.68 on the Test Set and 0.59 on the challenging Post-2021 Set, compared with 0.63 and 0.50 for DILI-Predictor, respectively. More importantly, we establish a baseline for mechanistic hypothesis generation, where HADES achieves a Hypothesis Alignment Fuzzy Jaccard Index of 0.16. This result underscores the inherent complexity of the task while highlighting the need for advanced explainable approaches in predictive toxicology.

cs.AI↗

arXiv2021

A topological selection of folding pathways from native states of knotted proteins

Understanding the biological function of knots in proteins and their folding process is an open and challenging question in biology. Recent studies classify the topology and geometry of knotted proteins by analysing the distribution of a protein's planar projections using topological objects called knotoids. We approach the analysis of proteins with the same topology by introducing a topologically inspired statistical metric between their knotoid distributions. We detect geometric differences between trefoil proteins by characterising their entanglement and we recover a clustering by sequence similarity. By looking directly at the geometry and topology of their native states, we are able to probe different folding pathways for proteins forming open-ended trefoil knots. Interestingly, our pipeline reveals that the folding pathway of shallow knotted Carbonic Anhydrases involves the creation of a double-looped structure, differently from what was previously observed for deeply knotted trefoil proteins. We validate this with Molecular Dynamics simulations.

q-bio.BM↗

arXiv2019

$θ$-curves in proteins

Apart from the knots formed by the main-chain, the proteins can form numerous topological structures, when included the covalent and ion-mediated interactions. In this work, we define the protein non-trivial $θ$-curves and identify 7 different topologies in all structures known up to date. We study the correlation of the motif with the function and organism of origin, and pointing the similarity with main-chain knots, we show that some motifs may indeed be functional. We also analyze the folding and bridge-induced stability of an exemplary protein with $θ$-curve motif and provide a catalogue of possible $θ$-curves in proteins.

cond-mat.soft↗