Big data, AI tech at the center of ALS Association partnership with GNS
ALS, Amyotrophic Lateral Sclerosis, is a progressive nervous system disease that affects nerve and muscle function. Currently, treatments can’t reverse the damage of ALS but progression of symptoms and prevent complications. There is no cure for ALS and it can eventually become fatal.
The ALS Association is providing funding to support GNS’ use of its artificial intelligence (AI) technology to create a disease model.
“GNS will receive and process data from the Answer ALS dataset which includes whole genome sequencing, proteomics, RNA sequencing, epigenomics, lipidomics, metabolomics and clinical information. GNS will then develop models using the clinical and molecular ‘omic data from patients samples,” GNS vice president of marketing, Patty Kellicker told us.
In response to the ice bucket challenge – which raised $115m for ALS research – Dr. Lucie Brujin, chief scientist at the ALS Association said, “one of the strategies was to see how the association could make a very significant impact and in particular engage and learn more through patients with the disease.”
One of these areas was the building up of big data or precision medicine.
Under the partnership, GNS’ machine learning platform, REFS will use Answer ALS patient datasets. Answer ALS will be collecting data from 1,000 individuals with ALS, including clinical, genetic, molecular, and biochemical information. This will yield thousands of petabytes of information, which will require analysis.
GNS in partnership with The ALS Association will use this information to create mechanistic models that will allow in silico experiments to be performed on the computer.
Discoveries made from the virtual experiments will be applied to wet lab experiments and then clinical studies, according to the association
The GNS Healthcare model and interface will be continuously refreshed and accessible to clinicians and scientists within the ALS research community.
Big data and better clinical trials
Brujin said it’s difficult to find therapies for ALS that are potentially useful for everyone with the disease, thus, the need for big data.
“The whole area of precision medicine is really new to the ALS field, so there are many groups now building data in this regard. But, so far, we have not really done the analysis of this data,” he explained.
It is thought that data mining of the datasets will provide information about ALS subpopulations, molecular pathways, and disease drivers. Additionally, Brujin believes that by using these tools they are being verified for use in other areas, and it will provide an opportunity for the research community to learn how big data can be useful, especially as it pertains to designing better clinical trials.
“Just like in the cancer field, doing the analysis started to enable us to segregate cancers into different types based on genetics…we’re trying to see that once you’ve got all this data and you’re starting to analyze it can we separate ALS into smaller groups that might define them in a better way, so we might be able to do clinical trials that are more successful,” Brujin said.
“I think this is the way forward, and I’m very excited for more partnerships.”