Clinical OMICS

MAY-JUN 2018

Healthcare magazine for research scientists, labs, pathologists, hospitals, cancer centers, physicians and biopharma companies providing news articles, expert interviews and videos about molecular diagnostics in precision medicine

Issue link: https://clinicalomics.epubxp.com/i/979295

Contents of this Issue

Navigation

Page 33 of 51

Data & Informatics 32 Clinical OMICs May/June 2018 www.clinicalomics.com Wave to Use Deep Genomics' AI Platform in Drug Discovery Collaboration Wave Life Sciences will use Deep Genom- ics' machine learning–driven biomedical platform to discover novel therapies for treating genetic neuromuscular disorders. The companies will test oligonucleotides against potential therapeutic targets within multiple genes implicated in neu- romuscular disorders. Wave will devel- op the new therapies using its propriety chemistry platform to validate targets and elucidate the implications of target inter- vention across different phenotypes, the companies said. The analysis of oligonucleotides will use Deep Genomics' machine learning plat- form to identify cause and effect relation- ships specific to neuromuscular-related targets that involve splicing regulation. Deep Genomics' platform is designed to discover drug candidates that target the genetic determinants of disease at the lev- el of RNA or DNA. The platform produces on-target and genome-wide off-target ef- fect data for every compound identified. Deep Genomics is using the platform to evaluate more than 69 billion molecules against 1 million targets, in silico. The ef- fort, dubbed "Project Saturn," is designed to generate a library of 1,000 compounds that are experimentally verified to manip- ulate cell biology as intended. n into a single unified search database in a way that cannot be accomplished through natural language processing. The GLP engine uses machine learning algorithms to continuously improve the quality of its genomic search results by intelligently elimi- nating false positive results from the indexing process, Genomenon said. The partnership will go beyond Genomenon's past collaborations, which have involved sharing of data with partners that had their own vari- ant curation/interpretation platforms: • Saphetor agreed in February with Genomenon to share genomic variant data between their software platforms. The deal links Mastermind with Saphetor 's VarSome.com knowledge base and aggregator for human genomic variants. • Also in February, GenomeNext said it embedded results from Mastermind within its Olym- pus genomic-analysis platform, through a partnership aimed at simplifying and accelerating variant curation for genomic analysis. • LifeOmic agreed in October 2017 to use Mastermind to automate prioritization of its genome sequencing data for clinical patients by annotating disease-causing variants with citations from relevant biomedi- cal literature. "Most of Genomenon's partner- ships provide valuable guidance on how Mastermind needs to deliver high quality search results that can be used to speed genomic interpretation in a clinical setting and test data to hone the search engine to deliver the high- est quality, clinically relevant search results for our partners," Klein said. As a development partner, he added, Veritas has provided direct input and guidance on the ACMG classification capabilities of Master- mind that will be helpful in acceler- ating their workflow. Veritas has also provided Genomenon with gold stan- dard datasets to tune its data prioriti- zation algorithms. As part of the companies' collab- oration, Veritas' clinical variant sci- ence team will advise Genomenon's product development team with ongoing requirements and product refinements. "Genomenon and Ver- itas are collaborating on using AI to improve the speed of genomic inter- pretation," Klein said. "By applying this computational intelli- gence to Mastermind's indexed data, we can identify the evidence needed to suggest ACMG interpretation for variants with supporting clini- cal evidence, automating the biggest bottleneck in the clinical sequencing workflow." (continued from previous page) Mastermind has indexed nearly 6 million scientific articles covering every disease, gene and variant, out of the 30 million titles & abstracts in PubMed. jxfzsy / Getty Images

Articles in this issue

Links on this page

Archives of this issue

view archives of Clinical OMICS - MAY-JUN 2018