GWU-Bioinformatics

A local, CPU-optimized pipeline for the automated extraction and evaluation of research papers. Built with Ollama and SQLite, this tool sifts through large-scale document sets to identify high-potential candidates for machine learning model development without external API costs.

LSTM-based time-series modeling of Continuous Glucose Monitoring (CGM) data the AI-READI dataset. The project extracts raw glucose trajectories, engineers circadian and variability features, and trains deep learning models to classify participants into metabolic health groups.

Codebase for BioPathNet to predict links in biomedical knowledge graphs