David Hoksza
Head of department, Associate professor
Major focus of David Hoksza is on the development of efficient algorithms in the area of structural bioinformatics and data visualization. He has participated in projects involving mostly protein and RNA structure with occasional excursions to the fields of cheminformatics (ligand-based virtual screening, exploration of chemical space), computational genomics (analysis of MinION data) and systems biology (visualization and analysis of molecular networks).
Courses
Data Visualization Techniques
NDBI042
Algorithms, databases and tools in bioinformatics
NDBI044
Project in bioinformatics
NPRG061
Latest publications
- Protein Language Models and Structure-Based Machine Learning for Prediction of Allosteric Binding Sites in Protein Kinases: An Explainable AI Framework Grounded in Energy Landscape-Encoded Frustration (2026)
- Predicting and Decoding Allosteric Binding Sites Using Protein Language Models and Structure-Based Machine Learning: An Energy Landscape-Guided Explainable AI Framework (2026)
- Beyond Exact Matches: Near-Hit Scoring for Protein Binding Site Prediction with Protein Language Models (2025)
- Hidden in protein sequences: Predicting cryptic binding sites (2025)
- CryptoBench: cryptic protein--ligand binding sites dataset and benchmark (2025)
- PrankWeb 4: a modular web server for protein--ligand binding site prediction and downstream analysis (2025)
- Unified visual-aware representations for data analytics (2025)
- Visualizations for universal deep-feature representations: survey and taxonomy (2024)
- Genomics 2 Proteins portal: a resource and discovery tool for linking genetic screening outputs to protein sequences and structures (2024)
- Cryptic binding site prediction with protein language models (2023)
- Visual representations for data analytics: user study (2023)
- PrankWeb 3: accelerated ligand-binding site predictions for experimental and modelled protein structures (2022)
- R2DT is a framework for predicting and visualising RNA secondary structure using templates (2021)
- Comprehensive characterization of amino acid positions in protein structures reveals molecular effect of missense variants (2020)
- PrankWeb: a web server for ligand binding site prediction and visualization (2019)
- MINERVA API and plugins: opening molecular network analysis and visualization to the community (2019)
- P2Rank: machine learning based tool for rapid and accurate prediction of ligand binding sites from protein structure (2018)
- MolArt: a molecular structure annotation and visualization tool (2018)
- TRAVeLer: a tool for template-based RNA secondary structure visualization (2017)
Maintainer of
MolArt
MolArt is a responsive, easy-to-use JavaScript plugin which enables users to view annotated protein sequence and overlay the annotations over a corresponding experimental or predicted protein structure.
P2Rank
P2Rank is a state-of-the-art machine learning-based method for ligand binding sites prediction based on protein structure.
Traveler
Traveler is an RNA sescondary structure visualization tool implementing a template-based approach enabling to lay out even the largest RNA structures in the standard orientation.