About
BioSymetrics is an AI-driven drug discovery company tackling one of the pharmaceutical industry's greatest challenges: the ~90% clinical trial failure rate largely attributed to poor target selection. By taking a phenomics-driven approach rather than reductionist methods, BioSymetrics integrates massive datasets—including over 77 million longitudinal patient records and more than 1 million genomic profiles—with machine learning models to identify and validate drug targets with greater accuracy and translational confidence. The platform bridges the gap between human disease phenotypes and model systems such as zebrafish, using computer vision and AI to analyze complex biological behaviors. Their KCC2 epilepsy program exemplifies this: computer vision algorithms extract 15,000+ data points per in vivo zebrafish video, identifying small molecules that reverse seizure phenotypes. In just four months, this approach identified novel hits including BIOS-1073. BioSymetrics partners with global pharmaceutical companies, top hospitals, health systems, and leading genomics organizations. Their current pipeline focuses on neurological, cardiometabolic, and rare diseases, with lead programs in common and rare epilepsies. The company is remote-first with offices in Toronto and New York, and is purpose-built for pharmaceutical partners and biotech organizations seeking to accelerate and de-risk drug discovery through AI and in vivo validation.
Key Features
- Phenomics-Driven Target Discovery: Connects disease phenotype and genotype data between humans and model systems using machine learning to identify high-confidence drug targets at scale.
- Massive Longitudinal Dataset: Leverages over 77 million longitudinal patient records and 1 million+ genomic profiles sourced through strategic partnerships with pharma, hospitals, and genomics organizations.
- AI-Powered Computer Vision: Analyzes 15,000+ data points per in vivo video using computer vision algorithms to detect phenotypic changes in model organisms such as zebrafish.
- In Vivo Validation Pipeline: Couples machine learning predictions with high-content in vivo validation to confirm target relevance before advancing programs, increasing translational confidence.
- Strategic Pharmaceutical Partnerships: Collaborates with global pharma companies, top hospitals, and genomics organizations to continuously expand its diverse and dynamic dataset.
Use Cases
- Pharmaceutical companies seeking AI-assisted target discovery to reduce the risk of clinical trial failure due to poor target selection
- Biotech firms looking to identify novel drug targets in neurological or rare disease programs using phenomics data
- Research organizations aiming to integrate multi-modal patient and genomic data for precision medicine development
- Drug discovery teams focused on epilepsy or cardiometabolic diseases seeking validated, high-confidence lead compounds through in vivo models
- Healthcare systems and genomics organizations partnering to contribute longitudinal patient data for translational drug research
Pros
- Addresses Root Cause of Trial Failures: Directly tackles the ~90% clinical trial failure rate by improving target selection confidence through integrative phenomics and AI-driven validation.
- Rich, Diverse Biological Dataset: Access to 77M+ longitudinal patient records and 1M+ genomic profiles provides an exceptionally robust foundation for machine learning predictions and target discovery.
- End-to-End Discovery Workflow: Combines data integration, ML prediction, and in vivo validation in a single pipeline, reducing handoff friction and compressing discovery timelines.
Cons
- Enterprise and Partner-Only Access: The platform is designed exclusively for pharmaceutical and biotech partners, making it inaccessible to independent researchers or small organizations.
- Narrow Current Disease Focus: Pipeline is presently concentrated on neurological, cardiometabolic, and rare diseases, limiting immediate applicability across broader therapeutic areas.
Frequently Asked Questions
What is phenomics-driven drug discovery?
Phenomics-driven drug discovery integrates comprehensive disease phenotype data (observable biological characteristics) with genotype data to understand disease biology more holistically, enabling better target identification than traditional single-pathway or reductionist approaches.
What diseases does BioSymetrics focus on?
BioSymetrics primarily focuses on neurological, cardiometabolic, and rare diseases. Their lead programs are in common and rare epilepsies, including the KCC2 epilepsy program targeting seizure disorders.
How large is the BioSymetrics dataset?
BioSymetrics leverages more than 77 million longitudinal patient records and over 1 million genomic profiles across its platform, sourced through partnerships with hospitals, pharmaceutical companies, and genomics organizations.
How does BioSymetrics use computer vision in drug discovery?
Their computer vision algorithms analyze in vivo model organism videos—such as zebrafish—extracting 15,000+ data points per video to identify small molecules that reverse disease phenotypes like seizures, enabling rapid hit identification.
How can pharmaceutical companies partner with BioSymetrics?
BioSymetrics partners with global pharmaceutical companies, hospitals, and genomics organizations to advance drug discovery programs and contribute to its growing dataset. Interested parties can reach out through the contact section on their website.
