Waterloo University researchers have developed a new system that could significantly accelerate detection of new drugs and reduce the need for expensive and long-lasting laboratory tests.
The new technology called Pattern to Knowledge (P2K) can predict binding bioseconds in seconds and potentially reduce bottlenecks in drug research.
P2K uses artificial intelligence (AI) to utilize deep knowledge from data instead of relying solely on classical machine learning.
"P2K is an exchanger player because of its ability to detect subtle protein associations tangled in complex physique environments and strongly predict interactions based only on sequence data," said Andrew Wong, professor, Systems Design Engineering and Founding Director, Sample Analysis Center and Machine Intelligence (CPAMI). "The ability to approach this profound knowledge of proven scientific results moves biology research going forward. P2K has the power to transform how data can be used in the future."
Although a large amount of biological sequence data was collected, extracting significant and useful knowledge was not easy. P2K algorithms solve this challenge by disentangling multiple associations to identify and predict the amino acid linkages that regulate protein interactions. Since P2K is much faster than existing biosensor analysis software, with almost 30% better accuracy, it could significantly accelerate detection of new drugs. Drawing information from the cloud database, P2K could predict how tumor proteins and potential cancer treatments would interact.
Although still in the early prototype phase, Professor Wong and his team have announced the availability of P2K system on the network researchers to begin identifying new interactions of the biological sequence.
"Placing this AI technology into the hands of biomedical researchers will bring immediate results that could be used for future scientific discoveries," said Antonio Sze-To, a research associate, Systems Design Engineering and P2K co-author.
Since it analyzes sequential data, the applicability of P2K is not limited to biomedical research. P2K could benefit from the financial industry by making useful associations and forecasts for smart commerce or the cybersecurity sector predicting the likelihood of possible cyber attacks.
Research work, "Sample of Knowledge: Deep Knowledge-Directed Machine Learning for Researched Residual and Remotion Interaction", was recently published in Nature's Scientific Reports.
Fujitsu Develops Technology for Predicting Biochemical Reactions, Clarifying Gene Disorder Mechanisms
Andrew K.C. Wong et al. Sample to Knowledge: Deep Knowledge-Directed Learning Machine for Surviving-Residual Interaction Predictions, Scientific Reports (2018). DOI: 10.1038 / s41598-018-32834-z
University of Waterloo
A new deep knowledge system AI could solve narrow throats in drug research (2018, November 7th)
taken over on November 7, 2018
This document is subject to copyright. Apart from any kind of fair for private study or research, no
part can be played without a written permission. Content is only available for informational purposes.