Research area in AI: Large-scale data and knowledge fusion in the life sciences. I developed a Bayesian systems-based approach to analyze causal and dependency relations at multiple levels. Using kernel methods and semantic technologies, we developed repositioning methods for drug discovery. Using active learning and ensemble methods, we developed novel genetic measurement methods for precision medicine. Currently, we are extending our methodologies towards active, collaborative learning to support the fusion of horizontally and vertically distributed data and knowledge at multiple partners.
RG-IPI-2019-TP13/017 (Richter): „Deep Priors For Drugs (De novo hatóanyagjelölt generálás nagy mennyiségű bioaktivitási információkat felhasználó mély megerősítéses tanulással”, 2 év: 2020-202, PI
MELLODDY project, H2020/IMI2, G.A.No.: 831472 — MELLODDY, title: MachinE Learning Ledger Orchestration for Drug DiscoverY, 2019-202, leader at the department
2015-2018, OTKA 112915, Decision Support and Intelligent Automation of Next-Generation Sequencing Workflows, collaborator, PI: Ákos Jobbágy
2016-2020, OTKA 119866, Bayesian, systems-based methods for analyzing large health data sets, principal investigator, PI: Péter Antal
2017-2019, Central Europe Leuven Strategic Alliance (CELSA): HIDUCTION: Privacy preserving data and knowledge fusion in personalized biomedicine, PI: Yves Moreau (K.U.Leuven)
2012-2014, TÁMOP-4.1.2.A/1-11/1-2011-0079, Active learning in biotechnology and bioinformatics, leader at the department
2008-2011, OTKA PD 76348, Bayesian methods for the generalized feature subset selection problem and their biomedical applications, principal investigator