Péter Antal

GNN4DM: A Graph Neural Network-based method to identify overlapping functional disease modules

GNN4DM identifies overlapping functional disease modules by integrating network topology and genomic data, outperforming existing methods and discovering novel multimorbidity modules.

Towards explainable interaction prediction: Embedding biological hierarchies into hyperbolic interaction space

This paper improves drug-target interaction prediction by using hyperbolic embeddings and hierarchy regularization to create an interpretable latent space.

Transfer learning in heterogeneous drug-target interaction predictions using federated boosting

This paper develops a federated boosting approach for drug-target interaction prediction, enhancing transfer learning in heterogeneous tasks with limited data.

Gene co-expression analyses of health(span) across multiple species

The paper identifies conserved healthspan-related gene modules across species using co-expression network analysis and meta-analysis of transcriptomic data.

Genetic effects on educational attainment in Hungary

The paper assesses polygenic scores for educational attainment in Hungarian and English populations, finding significant associations in Hungary but not in England.

Healthspan pathway maps in C. elegans and humans highlight transcription, prolifera-tion/biosynthesis and lipids

The paper compiles healthspan-related gene pathways in C. elegans and humans using literature review and molecular interaction data, identifying common themes like transcription and lipids.