Projects

GNN Theory & Methods

bias–variance · topology‑aware generalization · model selection

Statistical principles for graph neural networks and graph algorithms with interpretable, reproducible pipelines.

Canonical Correlation Analysis

sparse CCA · reduced rank regression

High‑dimensional estimators exploiting sparsity and graph structure; confidence measures for downstream decisions.

Topic Modeling for count data

spatial topic models · tensor topic modeling

High‑dimensional estimators exploiting sparsity and graph structure; confidence measures for downstream decisions.

Uncertainty Quantification

conformal prediction · bootstrap diagnostics

Coverage‑calibrated UQ for structured estimators and GNNs to support reproducible scientific claims.

Structured Estimation

high-dimensional data; graph total-variation penalty

High‑dimensional estimators exploiting sparsity and graph structure.

Machine Learning

Data augmentation

Spatial Transcriptomics & RNA Velocity

tissue graphs · anisotropic kernels

Expression and velocity as complementary signals; unsupervised GNN embeddings and spatial PCA reveal spatial programs.

Photosynthetic Microbes

Chlamydomonas · cyanobacteria

Multimodal integration for thermotolerance and phototaxis; linking omic profiles to stress phenotypes.

Microbial Communities

marine & host‑associated

Graph‑based latent variable models for community structure, accounting for spatial and environmental context.