Alina Bazarova

I am a researcher working at the intersection of machine learning, bioinformatics, and mechanistic modelling, with a focus on developing scalable and interpretable AI methods for biological systems. My work aims to connect molecular data with biological function and enable experimentally testable hypotheses through principled computational models.

A central theme of my research is the development of multi-modal representation learning methods for biological data, including protein modelling frameworks such as OneProt that integrate sequence, structure, and functional information. I am also interested in simulation-based inference and Bayesian approaches for mechanistic biological systems, where uncertainty and data efficiency are key challenges.

I am Principal Investigator on several research projects in machine learning for biological systems, spanning protein modelling, generative AI (OneProtGPT and ProtheraEGFR), and computational protein design. I am also a PI and the coordinator of the PROFOUND, a multi-institutional consortium within the Helmholtz Foundational Model Initiative.

I teach courses in Bayesian and simulation-based machine learning, including
Introduction to Bayesian Statistical Learning I and II, Introduction to Simulation-Based Inference, with a focus on probabilistic modelling, uncertainty quantification, and data-driven inference for complex biological systems.