Trained by Patients-on-a-Chip.

A radical new approach 
to predict drug safety.

“Quris is applying a disruptive bio-AI approach that my colleagues and I have previously used with dramatic success: we have analyzed the human genome, discovering hundreds of novel microRNA genes, more than all the universities in the world combined!

At Quris, we are now applying this approach to revolutionize drug development: we use Patients-on-a-Chip to generate a massive proprietary dataset that is automated, highly predictive, and uses classification algorithms to train the machine learning model to better predict which drug candidates will safely work in humans.”

Isaac Bentwich, MD, Founder & CEO, Genomics-AI Pioneer

Isaac Bentwich, MD

Founder & CEO

Genomics-AI Pioneer

Bio-AI Machine-Learning

Bio-AI Clinical Prediction Platform

Predicting which drug candidates will be safe and efficacious in humans is a formidable challenge. Traditional pre-clinical data, used or generated by other AI-pharma companies – lab, mice, genomics – is easily accessible, but very poorly predicts clinical safety and efficacy (remember that 89% of drugs fail in clinical trials, despite promising pre-clinical data!). Quris’s unique machine-learning approach using the Bio-AI Clinical Prediction Platform is radically different: By using a patented process to test known safe and unsafe drugs on miniaturized Patients-on-a-Chip with our automated, high-throughput system, data is generated, classified, and continuously re-trains the machine learning algorithm which is highly predictive of clinical safety and efficacy.


Our AI Chip-on-Chip platform (18 granted and pending patents) allows automated testing of thousands of drugs on miniaturized Patients-on-a-Chip, while next-generation nano-sensors allow for continuous monitoring of the responses from each miniaturized organ to these drugs. Our machine-learning classification algorithm is trained by the data continuously generated in this high-throughput system.

‘Patients on a Chip’
Stem-cell genomic diversity

Leveraging the Genomic Diversity of Stem Cells

The ability to train the machine learning model on a single Patient-on-a-Chip is limited. Our Bio-AI Clinical Prediction Platform leverages powerful AI to train the system on hundreds of stem cell-derived Patients-on-a-Chip, which reflect an extremely broad genomic diversity. This is made possible through our exclusive collaboration with the New York Stem Cell Foundation, the world leader in stem-cell automation.

Scalable Platform

Our AI Chip on Chip platform is uniquely scalable. Successfully tackling the complexities of clinical prediction requires machine learning models to routinely run thousands, and eventually millions, of biological Patients-on-a-Chip experiments for AI training.  Limitations of older organ-chip devices preclude the ability to run such experiments, however our patented platform is highly scalable and tightly integrated with the AI, enabling massive experiments at a small fraction of the cost.

Scalable Platform
Pipeline: Fragile-X drug Validates Platform

Pipeline: Fragile-X Drug Validates Platform

The tremendous potential of our platform to de-risk and drastically cut drug development time is evidenced by our preparation for clinical testing of our first drug developed with our Bio-AI Clinical Prediction Platform targeting Fragile-X Syndrome. Fragile-X Syndrome ($200M market) is the most common inherited cause of autism and intellectual disabilities worldwide and to date, big pharma drugs have repeatedly failed in clinical trials. Novartis and Roche tried and failed developing a drug for Fragile-X. Our drug is based on multiple top-tier peer-reviewed publications, and is the only drug that addresses the root cause of the disease and holds potential even to curing the disease altogether.

our partners:

The Hebrew University of Jerusalem
Tel Aviv University