Grigoriy Gogoshin

DRUMBEAT: Temporally resolved interpretable machine learning model for characterizing state transitions in protein dynamics

DRUMBEAT is a machine learning model that identifies time-resolved cooperative events in protein dynamics, revealing folding pathways and critical residues.

Temporally Resolved and Interpretable Machine Learning Model of GPCR Conformational Transition

This paper develops an interpretable machine learning model to identify residue communities involved in GPCR conformational transitions, with applications to dopamine receptor subtype-specific drug design.

Bayesian Network Structure Learning: The NewCalibrated Minimum Uncertainty Criterion andDiscriminative Power Evaluation Framework

This paper introduces a calibrated Minimum Uncertainty criterion for Bayesian network structure learning and evaluates scoring criteria's discriminative power using real biomedical data.

BaNDyT: Bayesian Network modeling of molecular Dynamics Trajectories

BaNDyT uses Bayesian network modeling to analyze molecular dynamics trajectories, providing data-driven insights into protein function and interactions.

Characterization of DNA extrusion scenarios underlying aggregate Hi-C patterns by a new computational framework

The paper introduces a computational framework to reconstruct DNA extrusion scenarios from aggregate Hi-C data and demonstrates how DNA extrusion explains directional biases and dynamic promoter-enhancer interactions.

Dissecting Response to Cancer Immunotherapy by Applying Bayesian Network Analysis to Flow Cytometry Data

The paper develops a Bayesian network analysis pipeline to identify immune signaling changes in GEA patients before and after immunotherapy, detecting signals missed by conventional methods.

Synthetic data generation with probabilistic Bayesian Networks

This study introduces a probabilistic Bayesian Network simulation framework to generate synthetic data that better aligns with real biological data and network models.

New analysis framework incorporating mixed mutual information and scalable Bayesian networks for multimodal high dimensional genomic and epigenomic cancer data

The paper introduces a two-stage framework using mixed mutual information and scalable Bayesian networks to analyze multimodal cancer data, identifying 11 candidate genes with 6 validated in literature.