Konstantin Avchaciov

HARVEST extracts structured bioactivity data from pharmaceutical patents using AI, producing millions of records and revealing gaps in existing databases.

The paper tracks aging in sled dogs using a variational autoregressive model to show that lamivudine temporarily decelerates DNA methylation clocks but not long-term biological age.

ProtoBind-Diff generates novel, chemically valid ligands based on protein sequences without structural data, performing well in docking and binding affinity evaluations.

The paper uses graph neural networks to predict polypharmacological compounds that extend lifespan in C. elegans, finding 70% of selected compounds significantly increase lifespan.

This paper develops a deep learning model to identify a blood-based biomarker of aging in mice using phenotypic data, finding it correlates with aging hallmarks and predicts lifespan.

The paper analyzes longitudinal blood marker data to show aging is associated with declining physiological resilience and predicts a lifespan limit around 120-150 years.

The paper quantifies biological age and frailty using locomotor activity data from NHANES and UKB, linking aging trajectories to mortality and health outcomes.