Integration of multi-omics data for predictive modelling of drug efficacy in cerebral organoids

To date, using only electrophysiological data, Epiloid has achieved a classification accuracy in the low-90% in classifying drugs on their predicted ability to modulate hyperexcitability (as seen in epilepsy). This project seeks to enhance this accuracy by incorporating multi-omic datasets into Epiloid’s machine learning pipeline, which stands to dramatically assist in collecting a holistic picture of how new therapeutics affect neural circuitry of brain organoids. This provides a useful translational tool towards personalized medicine.
Epiloid Biotechnology is pioneering the integration of human-derived 3D cerebral organoids with machine learning (ML) to enhance preclinical drug discovery for neurological diseases. This project aims to bridge the gap between complex transcriptomic readouts and ML-driven drug efficacy predictions, leveraging human-relevant in vitro models. In collaboration with the Stem Cell Network, this project will support the ongoing R&D efforts funded through Epiloid’s Ontario Genomics BioCreate grant, which focuses on
integrating multi-omics data into predictive models for neurological drug development. The project includes the development of human iPSC cultures, cerebral organoid generation, and optimization of of SOPs for transcriptomic data preprocessing for ML model integration, and validation of these models using publicly available datasets.
The project will deliver several critical outcomes, including optimized SOPs for transcriptomic data generation, high-quality organoid datasets, and enhanced ML models that integrate electrophysiological and transcriptomic data. These advancements will position Epiloid as a leader in human-relevant preclinical testing, directly supporting its long-term goal of becoming a key partner in precision drug discovery.Furthermore, this partnership aligns closely with SCN’s national mandate to accelerate the development of
regenerative medicine and cellular therapies through collaborative research.
Anticipated benefits include the improved prediction of drug efficacy and safety, reducing costly late-stage failures; accelerated drug discovery timelines through better preclinical models; and enhancing Epiloid’s competitive advantage through proprietary organoid and machine learning data pipelines.

Faculty Supervisor:

Jeremy Sivak

Student:

Partner:

Stem Cell Network;Epiloid Biotechnology Inc.

Discipline:

Life Sciences

Sector:

Professional, scientific and technical services

University:

University of Toronto

Program: