Back to Table of contents

Primeur weekly 2017-04-18

Quantum computing

QxBranch and Commonwealth Bank Australia launch quantum computing simulator ...

Indistinguishable photons key to advancing quantum technologies ...

Recent advances and new insights into quantum image processing ...

Focus on Europe

Teratec 2017 Forum issues Call for Participation ...

Hazel Hen helps explain ultrafast phase transition ...

Hardware

Engility to pursue NASA advanced computing services opportunity ...

DDN names Jessica Popp General Manager of IME business unit ...

Eni fires up its HPC3, the new hybrid high performance computer for E&P activities ...

DDN advances object storage performance and delivers industry's most flexible and cost-effective data protection ...

Asetek to receive RackCDU D2C order for new HPC installation ...

PSNC deploys ADVA Optical Networking 96-channel 100G core solution in pan-European research network ...

Putting a spin on logic gates ...

Tool for checking complex computer architectures reveals flaws in emerging design ...

System better allots network bandwidth, for faster page loads ...

Applications

SDSC to enhance campus research computing resources for bioinformatics ...

U.S. Department of Energy's INCITE programme seeks advanced computational research proposals for 2018 ...

Tutorials schedule announced for PEARC17 ...

Fujitsu awarded three prizes for science and technology from MEXT ...

Fujitsu and Grid partner to jointly develop AI services ...

IBM brings Anaconda Open Data Science platform to IBM Cognitive Systems ...

Jefferson Lab scientists eavesdrop on chatter of sub-atomic world ...

Buckle up - Climate change to increase severe aircraft turbulence ...

Beyond the frontiers of Supercomputing ...

Scientists develop a novel algorithm, inspired on the behaviour of bee colonies, which will help dismantling criminal social networks ...

The Cloud

Atos leads C2NET consortium - the first collaborative Cloud-based platform for SMEs to support manufacturing management ...

Comcast Business now provides enterprises with dedicated links to IBM Cloud ...

Nimbix ushers in next-generation GPUs for Cloud-based deep learning ...

USFlash

Group works toward devising topological superconductor ...

Stanford researchers create deep learning algorithm that could boost drug development ...

Biased bots: Human prejudices sneak into artificial intelligence systems ...

Stanford researchers create deep learning algorithm that could boost drug development

Stanford chemistry Professor Vijay Pande and his students see a future for machine learning in the early stages of drug development. Credit: L.A. Cicero3 Apr 2017 Stanford - Artificially intelligent algorithms can learn to identify amazingly subtle information, enabling them to distinguish between people in photos or to screen medical images as well as a doctor. But in most cases their ability to perform such feats relies on training that involves thousands to trillions of data points. This means artificial intelligence doesn't work all that well in situations where there is very little data, such as drug development.

Vijay Pande, professor of chemistry at Stanford University, and his students thought that a fairly new kind of deep learning, called one-shot learning, that requires only a small number of data points might be a solution to that low-data problem.

"We're trying to use machine learning, especially deep learning, for the early stage of drug design", stated Vijay Pande. "The issue is, once you have thousands of examples in drug design, you probably already have a successful drug."

The group admitted the idea of applying one-shot learning to drug design problems was farfetched - the data was likely too limited. However, they'd had success in the past with machine learning methods requiring only hundreds of data points, and they had data available to test the one-shot approach. It seemed worth a try.

Much to their surprise, their results, published April 3 inACS Central Science, show that one-shot learning methods have potential as a helpful tool for drug development and other areas of chemistry research.

Other researchers have successfully applied one-shot learning to image recognition and genomics, but applying it to problems relevant to drug development is a bit different. Whereas pixels and bases are fairly natural types of data to feed into an algorithm, properties of small molecules aren't.

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.

"We worked on some prototype algorithms and found that, given a few data points, they were able to make predictions that were pretty accurate", stated Bharath Ramsundar, who is a graduate student in the Vijay Pande lab and co-lead author of the study.

However, Bharath Ramsundar cautioned that this isn't a "magical" technique. It was built off of several recent advances in a particular style of one-shot learning and it works by relying on the closeness of different molecules, as indirectly indicated by their formula. For example, when the researchers trained their algorithm on the toxicity data and tested it on the side effect data, the algorithm completely collapsed.

People concerned about AI taking jobs from humans have nothing to fear from this work. The researchers envision this as groundwork for a potential tool for chemists who are early in their research and trying to choose which molecule to pursue from a set of promising candidates.

"Right now, people make this kind of choice by hunch", Bharath Ramsundar stated. "This might be a nice compliment to that: an experimentalist's helper."

Beyond giving insight into drug design, this tool would be broadly applicable to molecular chemistry. Already, the Vijay Pande lab is testing these methods on different chemical compositions for solar cells. They have also made all of the code they used for the experiment open source, available as part of the DeepChem library.

"This paper is the first time that one-shot has been applied to this space and it's exciting to see the field of machine learning move so quickly", Vijay Pande stated. "This is not the end of this journey - it's the beginning."

Han Altae-Tran, Massachusetts Institute of Technology, is also lead author on this paper. Aneesh S. Pappu, Stanford University, is co-author. Vijay Pande is also a member of Stanford Bio-X and the Stanford Child Health Research Institute; and a fellow at Stanford ChEM-H.

This research was funded by the Fannie and John Hertz Foundation.
Source: Stanford University

Back to Table of contents

Primeur weekly 2017-04-18

Quantum computing

QxBranch and Commonwealth Bank Australia launch quantum computing simulator ...

Indistinguishable photons key to advancing quantum technologies ...

Recent advances and new insights into quantum image processing ...

Focus on Europe

Teratec 2017 Forum issues Call for Participation ...

Hazel Hen helps explain ultrafast phase transition ...

Hardware

Engility to pursue NASA advanced computing services opportunity ...

DDN names Jessica Popp General Manager of IME business unit ...

Eni fires up its HPC3, the new hybrid high performance computer for E&P activities ...

DDN advances object storage performance and delivers industry's most flexible and cost-effective data protection ...

Asetek to receive RackCDU D2C order for new HPC installation ...

PSNC deploys ADVA Optical Networking 96-channel 100G core solution in pan-European research network ...

Putting a spin on logic gates ...

Tool for checking complex computer architectures reveals flaws in emerging design ...

System better allots network bandwidth, for faster page loads ...

Applications

SDSC to enhance campus research computing resources for bioinformatics ...

U.S. Department of Energy's INCITE programme seeks advanced computational research proposals for 2018 ...

Tutorials schedule announced for PEARC17 ...

Fujitsu awarded three prizes for science and technology from MEXT ...

Fujitsu and Grid partner to jointly develop AI services ...

IBM brings Anaconda Open Data Science platform to IBM Cognitive Systems ...

Jefferson Lab scientists eavesdrop on chatter of sub-atomic world ...

Buckle up - Climate change to increase severe aircraft turbulence ...

Beyond the frontiers of Supercomputing ...

Scientists develop a novel algorithm, inspired on the behaviour of bee colonies, which will help dismantling criminal social networks ...

The Cloud

Atos leads C2NET consortium - the first collaborative Cloud-based platform for SMEs to support manufacturing management ...

Comcast Business now provides enterprises with dedicated links to IBM Cloud ...

Nimbix ushers in next-generation GPUs for Cloud-based deep learning ...

USFlash

Group works toward devising topological superconductor ...

Stanford researchers create deep learning algorithm that could boost drug development ...

Biased bots: Human prejudices sneak into artificial intelligence systems ...