Bharath Ramsundar

Lognormal Neural Point Process Models for Interpretable Heartbeat Dynamics

The paper applies a lognormal mixture neural temporal point process to model heartbeat dynamics, showing improved robustness to noise compared to an inverse Gaussian model.

Towards a Physiological Scaling Law: Model Quality vs. Cohort Size for Stochastic Sequence Data

The paper evaluates scaling laws for stochastic sequence data using distributional calibration metrics, showing that performance improves with cohort size beyond point accuracy saturation.

Evaluating Long-Range Temporal Structure in Foundation Model-Based Forecasts of Heartbeat Dynamics

This study evaluates how well Time-Series Foundation Models predict heartbeat dynamics and finds they fail to capture long-range temporal dependencies, leading to increasing prediction errors over time.

ProteomeScan: A Toolkit For Target Validation By Proteome-Wide Docking And Analysis

ProteomeScan identifies protein targets for drug-like molecules through proteome-wide docking and analysis, showing improved accuracy over random baselines.