New AI tech reduces research bottlenecks
Antonio Sze-To, research associate at the University of Waterloo and co-inventor of P2K told us that the platform was developed to predict interacting sites between bio-sequences without 3D structures, making it an economical and efficient option for preclinical research.
Sze-To said, “Biosequence interaction is important since it reveals cell fate and the molecular basis of diseases.” Identifying the interacting sites between bio-sequences was traditionally a time-consuming and expensive process involving extensive wet-lab experiments.
“Nowadays, the prediction of binding of bio-sequences is based on the close contact data procured from existing three-dimensional interaction complexes,” he explained. “However, such closed contacts sometimes are caused by other physio-chemical and stereo factors in complex interacting environments. Hence such statistics could be masked and misleading, impairing the prediction.”
P2K has algorithms supported by artificial intelligence (AI) that are capable of disentangling multiple associations to identify and predict amino acid bindings. It is able to provide deeper knowledge of reflecting residue-residue interactions and their specific functional spaces unaffected by the environment, according to Sze-To.
The system has been implemented online and is available to the public. Sze-To said, “Putting this AI technology in the hands of biomedical researchers will generate immediate results, which could be used for future scientific discoveries.”
Researchers explained that since the technology analyzes sequential data its use isn’t limited to biomedical research. It could be applied to the financial industry by making associations and predictions for smart trading.
P2K is just one of the ways in which AI has branched into preclinical research. Sze-To told us that $1.2bn (€1bn) has been invested in establishing AI startup companies.
Researchers are using AI to catalyze drug discovery by improving predictive models for efficient design and optimized multi-drug regimens, according to a recent report published by PreScouter.
The report said AI has the potential to revolutionize the current timescale and scope of clinical discovery and development.
Bluebird and Gritstone Oncology collaborated on an AI platform to develop and commercialize cancer treatments. Additionally, Evotec invested €15m to further its AI-driven drug discovery, Owkin invested €9.6m in its AI platform, and Insilico and Juvenescence entered a joint venture to develop AI-discovered molecules.