How AI algorithms could help design new drugs - Futurity | MyThinkPond

[![](/img/2017/04/ai.jpg)](/img/2017/04/ai.jpg)A new kind of AI algorithm—designed to work with a small amount of data—may be able to assist in the early stages of drug development. Artificial intelligence doesn’t work all that well in situations where there is very little data, such as drug development. A new technique called one-shot learning, that requires only a small number of data points might be a solution to that low-data problem. To make molecular information more digestible, the researchers first represented each molecule in terms of the connections between atoms (what a mathematician would call a graph). This step highlighted intrinsic properties of the chemical in a form that an algorithm could process. With these graphical representations, the group trained an algorithm on two different datasets—one with information about the toxicity of different chemicals and another that detailed side effects of approved medicines. From the first dataset, they trained the algorithm on six chemicals and had it make predictions about the toxicity of the other three. Using the second dataset, they trained it to associate drugs with side effects in 21 tasks, testing it on six more. In both cases, the algorithm was better able to predict toxicity or side effects than would have been possible by chance.

Article: http://www.futurity.org/one-shot-learning-algorithm-1394022/

Learn more  One-Shot Learning

Venkatt Guhesan avatar
About Venkatt Guhesan
Enterprise Software Architect currently working at DataDirect Networks developing a highly scalable monitoring solution called DDN Insight. My passions include all things Open Source including Linux (CentOS, Red Hat, Gentoo, Open Stack, Docker & Ubuntu), embedded ARM processors and microcontrollers (Arduino, UDOO, Cubieboard, BeagleBoard, OnionIO, TI's Zigbee). Coming from an Electrical and Computer Engineering background, working in developing and designing enterprise - distributed, highly scalable software requires a unique set of software skills with a good understanding of the hardware. This makes his work challenging and interesting. In his free time, he spends his time with his family and two children. He also enjoys researching on topics including - world ancient history and yoga. Visit the 'About' section for more details.
comments powered by Disqus