About QuAD3 Platform

About QuAD3 Platform
About QuAD3 Platform

Leveraging Our QuAD³ Platform for Clinical Development and Novel Drug Discovery

Our proprietary QuAD³ platform, powered by our AI and ML capabilities, allows us to extract deep phenotypic and genomic data that defines novel biological insights from our patient cohorts. We plan to apply our QuAD3 platform across a range of biologically relevant queries to provide us with additional context around the role the adaptive stress response and innate and adaptive immunity play in treatment resistance, cancer relapse and metastasis. We believe that such insights will, in turn, inform our ongoing efforts in adaptive stress target discovery, patient enrichment, and clinical development of our innovative therapeutic programs.

Applying Our QuAD³ Platform to Strengthen Our Pipeline


We are using our QuAD3 platform to understand how patient outcomes correlate with the adaptive stress response.  We are pioneering the understanding of the adaptive stress response and we believe that AI/ML will be an important tool in elucidating this complex biology.


Clinical trials can fail for a variety of reasons. We believe applying our QuAD3 platform can help us to better design our clinical trials by identifying effective biomarkers that can help optimize patient selection and enrichment strategies, and to prioritize clinical indications where stress modulation would have the most meaningful impact.


We are also applying our Artificial Intelligence/Machine Learning capabilities to drive novel stress pathway-related target identification and in-silico validation as a means of expanding our clinical pipeline of novel cancer therapies.

The Importance of Integrating Both Molecular and Clinical Data Sources

We believe it is critical to integrate multi-omic data sets to inform our understanding of the role of adaptive stress in cancer progression.

Despite the complexity of multi-omics data, our QuAD3 platform allows us to integrate and deconvolute genomic and multi-omic data sets, which has allowed us to develop a preliminary 30 gene stress signature, which was supported which was supported by our in vivo signal cell RNA sequencing, or RNA seq, data.

Differentiated Computational Tools to Develop a Deep Understanding of the Drivers of Disease

Our QuAD3 platform integrates multidisciplinary inputs to generate insights to guide our clinical development. Our platform takes a statistics-based ensemble-AI approach to help determine causality and improve replicability of findings. This innovative computational platform uses multiple AI/ML algorithms to incorporate different data types, cluster data without losing information, generate novel insights, and tie them back to existing literature. We believe this approach will allow us to refine our understanding of the complexity that underscores adaptive stress in cancer.

Our Key Capabilities and Strategies

Biomarker Discovery

Biomarker Discovery

Algorithms utilized to allow us to better project patient response to our therapeutic candidates

qML Computing Strategies

qML Computing Strategies

Machine learning strategies capable of superior performance on smaller, Phase 2 clinical trial-sized datasets

Ensemble AI Approach

Ensemble AI Approach

Multiple algorithms applied in sequence which is an approach that is intended to improve performance and ensure replicability of findings

In conjunction with our tools, we leverage collaboration with leading institutions across the US and Europe that are at the forefront of advancements in AI/ML for the biomedical sciences.

Our AI/ML team has a documented track record of publications identifying disease biology driver mechanisms.

Our QuAD3 platform was built upon these experimentally validated capabilities in collaboration with multiple top tier institutions. We have published our discoveries and novel methods in TIER-1 peer-reviewed journals.

Below is limited selection of our recent publications, including the first successful application of quantum ML approaches to human multi-omics cancer datasets.