search.noResults

search.searching

dataCollection.invalidEmail
note.createNoteMessage

search.noResults

search.searching

orderForm.title

orderForm.productCode
orderForm.description
orderForm.quantity
orderForm.itemPrice
orderForm.price
orderForm.totalPrice
orderForm.deliveryDetails.billingAddress
orderForm.deliveryDetails.deliveryAddress
orderForm.noItems
LABORATORY INFORMATICS


AI advances healthcare research


SOPHIA KTORI EXPLORES THE ROLE OF AI AND DEEP LEARNING IN HEALTHCARE – IN THE FIRST OF A TWO PART SERIES


Harnessing AI for drug discovery applications will significantly speed the identification of


promising drug candidates, believes Matt Segall, CEO at Optibrium. The UK-based firm, together with partners Intellegens and Medicines Discovery Catapult, recently received a grant from Innovate UK to help fund a £1 million project focussed on combining Optibrium’s existing StarDrop software for small molecule design, optimisation and data analysis, and Intellegens’ deep learning platform Alchemite. The aim is to develop a novel, deep learning AI-based method for predicting the ADMET (absorption, distribution, metabolism, excretion and toxicity) properties of new drugs candidates. Ultimately, the platform could help to guide the selection and design of more effective, safer compounds earlier in the discovery process, Segall states. ‘The pharma industry is increasingly


leveraging quantitative structure-activity relationship [QSAR] models to help predict the biological activity and toxicity of drug compounds, based on their structure,’ he explains. ‘People have been applying deep learning techniques to datasets to build these QSAR models, but frankly, they often add very little new insight. Published data suggests that using conventional deep learning algorithms to build QSAR models adds very little extra intelligence, compared with random forest and other algorithms that are commonly used.’ The reason for this is that deep learning methods are designed to deal with very large, but complete datasets, whereas the pharmaceutical sector commonly deals with very sparse and incomplete data, Segall continues.


www.scientific-computing.com | @scwmagazine


“The platform combines two-dimensional and three-dimensional structure activity relationships with de novo design, to help explore new strategies for optimisation”


‘A big pharma company might have one


to two million compounds and thousands of assays, but there’s no way that every compound will have been through every type of assay. Their overall pot of data may only be a few per cent complete and most of the information that they would ideally have for every compound, or for every assay, is actually missing. Conventional deep learning methods really can’t work with that sort of data input,’ said Segall. The Intellegens’ platform is designed to be able to bridge these gaps, by learning underlying correlations and relationships between different bioactivities and different assay endpoints, so that missing properties, relationships and activities can be predicted. ‘The platform is a unique approach to deep learning, based on sparse and uncertain data, but it’s also very generic and can be applied in almost any vertical,’ Segall notes. ‘Intellegens’


algorithms are already being exploited with commercial success in the materials science space.’ Optibrium’s own StarDrop software is a very comprehensive platform for small molecule design optimisation and data analysis. ‘StarDrop combines conventional data visualisation techniques with a unique decision analysis algorithm to tackle multi-parameter optimisation challenges, and also a very comprehensive suite of computational chemistry capabilities for understanding the structure activity relationships within existing projects, and applying those to guide the design of new compounds,’ Segall comments. ‘The platform combines two- dimensional and three-dimensional structure activity relationships with de novo design, to help explore new strategies for optimisation. Our platform is an ideal fit with Intellegens’ deep learning capabilities for the compound optimisation space.’ The third partner in the funded project, Medicines Discovery Catapult, is providing specialist expertise in data curation, to increase the amount of data available to the algorithms.


New avenues for research ‘We have some data that we have generated over the years, and the companies who we are working with have big datasets of their own based on their


October/November 2018 Scientific Computing World 21


g


Panuwatccn/ WhiteMocca/Shutterstock.com


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36