Vira Regen - Building Vira Regen’s AI Engine for Bioelectronic Medicine

Company Overview

Vira Regen is a Salt Lake City–based regenerative medicine company developing noninvasive bioelectronic medicines through electrogenetics, a new class of therapy that uses precisely engineered electrical signals to modulate gene expression on demand and drive tissue regeneration. The central scientific insight behind the company is that electrical stimulation parameters — waveform, frequency, amplitude, duty cycle, and exposure duration — act much like a dosing regimen, producing distinct and reproducible changes in gene expression depending on how they are combined. Our lead program, OsteoVolt™, applies this approach to fracture healing and showed approximately 50% faster healing in preclinical models while advancing along an FDA Class II 510(k) pathway, with a pipeline spanning knee osteoarthritis, osteoporosis, cancer electrotherapy, and human performance and longevity. The company is preclinical stage and has received a $2 million non-dilutive award through the DoW-MTEC consortium.

Project Description

Students on this project will help build Data Regen, Vira Regen’s AI engine for curating more accurate and effective stimulation therapies. The core task is to design and prototype a database that links electrical stimulation parameters to gene and sequence data, so that the relationship between a stimulation protocol and the biological response it produces becomes queryable rather than anecdotal. The team will design a schema capable of representing stimulation protocols, experimental conditions, and transcriptomic and genomic readouts; build ingestion and normalization pipelines that bring in public datasets such as GEO and related repositories alongside Vira Regen’s internal preclinical data; and apply customized machine learning — including supervised models that predict gene expression response from stimulation parameters and unsupervised methods that cluster co-responsive gene cascades — to surface candidate protocols worth testing at the bench. Attention to reproducibility, data versioning, and documentation is as important as model accuracy, since this system is intended to guide real laboratory work. Deliverables are a working database and pipeline, trained baseline models with a documented validation approach, a simple query or dashboard interface for the scientific team, and a technical roadmap describing what it would take to extend Data Regen toward closed-loop therapy design.

Student Alignment and Interests:

Data Engineering · Data Analytics · Software Development