CAMBRIDGE, Mass.—Bristol-Myers Squibb Co. (BMS) has entered into a new venture with GNS Healthcare that will see the companies use genetic, clinical and molecular data with super-computer analytics to identify biomarkers of immune-inflation.
The agreement will allow Cambridge, Mass.-based GNS to use information from a BMS clinical trial while using its own supercomputer-driven reverse-engineering and forward-simulation (REFS) platform for creating computer models to identify key molecular mechanisms.
REFS is licensed to GNS Healthcare from parent company Via Science.
According to Thomas A. Neyarapally, senior vice president of corporate development for GNS Healthcare, this is the first time the two companies have worked together. The goals of the collaboration, according to Neyarapally, are to "learn novel drug and disease biology in the form of computer models that enable virtual clinical trials that will simulate the clinical effect of inhibiting various drug targets, and to help predict novel molecular targets to combat immuno-inflammation that would be effective for specific types of patients."
The project will involve GNS using patient data comprising genetic information, specific blood markers, gene expression levels and clinical results to construct a comprehensive disease model that it is hoped will support future trials. The data are being used from a BMS drug clinical trial.
"REFS is comprised of integrated machine learning algorithms and software that extract 'causal' relationships from complex, multi-dimensional data and enable the simulation of billions of 'what if?' hypotheses to explore novel unseen conditions and predictions forward in time," says Neyarapally.
Neyarapally notes that REFS enables the "hypothesis-free" analysis of very complex systems by building models directly from experimental and clinical data, freeing researchers from the constraints of working within the realm of "known" biology —which he says is often incorrect and usually incomplete.
"Most other offerings do not explicitly connect genetic to multiple molecular to clinical data modalities in one simulation model," he adds.
The companies predict the venture could help to predict novel molecular targets to battle immune-inflation that would be effective for specific types of patients.
"This is the first step in building a significantly advanced in-silico research paradigm that may enable development of new therapies based directly on human data and address the needs of patients with particular genetics and molecular phenotypes, ultimately enabling the optimization of individualized patient outcomes," says Iya Khalil, executive vice president and co-founder of GNS Healthcare.
Biomarkers identified by model simulation will be validated, after which BMS will determine how it wants to utilize the discoveries in future drug discovery and development efforts.
While GNS hasn't revealed the financial terms of the agreement with BMS, it could provide another level of validation for its REFS technology.
The REFS approach has already delivered results in a rheumatoid arthritis study involving data from Biogen Idec, and last month the National Cancer Institute's Center for Advanced Preclinical Research hooked up with GNS for its research of lung cancer.
"A successful collaboration," Neyarapally says, "would entail validation of the predictive power of the models built in the collaboration and the use of the discoveries in future drug discovery and development efforts."
GNS is also currently collaborating with Brigham and Women's Hospital (BWH) on asthma patient therapies, among other partnerships.
