{"id":31514,"date":"2017-03-15T07:58:22","date_gmt":"2017-03-15T07:58:22","guid":{"rendered":"http:\/\/electronicsmaker.com\/?p=31514"},"modified":"2017-03-15T07:58:22","modified_gmt":"2017-03-15T07:58:22","slug":"how-ai-promises-to-speed-drug-development","status":"publish","type":"post","link":"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development","title":{"rendered":"How AI Promises to Speed Drug Development"},"content":{"rendered":"

Courtesy: NVIDIA<\/span><\/em><\/span><\/p>\n

\n

Creating New Drugs, Faster<\/em><\/h3>\n<\/div>\n

Alzheimer\u2019s. Cancer. Parkinson\u2019s. These devastating diseases are just a few of the hundreds that scientists are struggling to cure with new medicines in the face of soaring drug-discovery costs and testing times.<\/p>\n

Now, artificial intelligence is showing the potential to be a faster, more efficient way to find and develop new drugs. A growing number of companies and university researchers are tackling some of medicine\u2019s toughest problems by using AI computing to predict which drug molecules are most likely to be effective treatments.<\/p>\n

\u201cThis is a revolution in the pharmaceutical industry,\u201d said Alex Zhavoronkov, CEO of Insilico Medicine, which uses GPU-accelerated deep learning to target cancer and age-related illnesses.<\/p>\n

Drug Discovery Now: It\u2019s Broken<\/b><\/h2>\n

The revolution can\u2019t come soon enough. Developing a new drug costs an average of nearly $2.6 billion and can take as long as 14 years, according to figures from a Tufts University study and the U.S. Food and Drug Administration.<\/p>\n

Even then, less than 10 percent of potential medicines make it to market, said Jackie Hunter, CEO of BenevolentBio, the life sciences arm of London\u2019s BenevolentAI.<\/p>\n

\u201cThe economics are not sustainable,\u201d she said \u201cWe need to think about doing things differently.\u201d<\/p>\n

AI Drug Discovery<\/b><\/h2>\n

BenevolentBio aims to reinvent drug discovery by using deep learning and natural language processing to understand and analyze vast quantities of bioscience information \u2014 patents, genomic data and the more than 10,000 publications uploaded daily across all biomedical journals and databases.<\/p>\n

\u201cHumans alone cannot process all this information to advance scientific research,\u201d Hunter said.<\/p>\n

BenevolentBio\u2019s deep learning software, powered by the NVIDIA DGX-1 AI supercomputer, ingests and analyzes the information to find connections and propose drug candidates.<\/p>\n

Hunter\u2019s team also uses deep learning to generate molecules with properties that are likely to be effective against specific diseases, such as new molecules for neurodegenerative diseases, but without adverse cardiac or liver side effects.<\/p>\n

So far, BenvolentBio has identified two drugs for amyotrophic lateral sclerosis (better known as ALS) \u2014 a progressive neuro-degenerative disease \u2014 that in its tests worked better than those on the market, Hunter said. After it obtains patents, she said BenevolentBio will approach pharmaceutical companies to pursue development.<\/p>\n

\"Cells<\/a>
Cells believed to play a role in the progression of ALS. BenevolentBio used deep learning to find two possible treatments. Image courtesy of Oregon State University.<\/figcaption><\/figure>\n

Imagining a Molecule<\/b><\/h2>\n

Today, pharmaceutical companies must screen large numbers of molecules for drug candidates, then painstakingly test each in the hope of finding a winner. This expensive process takes enormous amounts of time.<\/p>\n

Baltimore\u2019s Insilico Medicine is working to flip this dynamic on its head by \u201cimagining\u201d new cancer-like molecules with specified properties. They do this with a relatively new deep learning technique known as a generative adversarial network (GAN).<\/p>\n

GANs can create entirely new data that is indistinguishable from real data by using two competing neural network models. A generative model tries to generate output that \u201clooks like\u201d real data. A discriminative model takes input from both the generative model and real data and tries to distinguish between them.<\/p>\n

Generative models are commonly used to create images, speech or text, but this is the first time a GAN has been used for cancer drug discovery, Insilico said.<\/p>\n

Molecules on Demand <\/b><\/h2>\n

In their recent paper, researchers in Insilico\u2019s Pharma AI division describe how their network used historical biological and chemical data to\u00a0imagine 69 new molecules with the potential to fight cancer.<\/p>\n

\u201cRather than using trial and error when looking for molecular leads, you can tell the network to generate those leads and it will do it on demand,\u201d said Insilico\u2019s Zhavoronkov. Insilico is now extending this work to imagine antimicrobial leads.<\/p>\n

Insilico trained its algorithms using the CUDA\u00a0parallel computing platform, Tesla K80 accelerators and TITAN X GPUs with cuDNN\u00a0on the Theano and TensorFlow deep learning frameworks.<\/p>\n

\"Developing<\/a>
Developing a new medicine is more expensive and time-consuming than ever. New methods using AI could change that.<\/figcaption><\/figure>\n

Molecules by Design<\/b><\/h2>\n

Al\u00e1n Aspuru-Guzik, a Harvard University professor of chemistry and chemical biology, is also seeking to build treatments from the ground up using a deep learning neural network for what he calls \u201cinverse molecular design.\u201d<\/p>\n

To find candidates for drug development, the deep learning software proposes novel molecules that posses\u00a0certain properties, Aspuru-Guzik said. The software also generates new molecular structures by combining properties of existing drugs.<\/p>\n

These methods might accelerate drug discovery, Aspuru-Guzik said. He describes his research using a generative model for\u00a0automatic chemical design in a recent paper published in ArXiv, the online archive of research articles.<\/p>\n

Aspuru-Guzik trained his deep learning software on 250,000 drug-like molecules using the CUDA parallel computing platform, NVIDIA Tesla K80 GPU accelerators and cuDNN with the Theano deep learning framework.<\/p>\n

No one expects AI drug discovery to replace scientists. Most believe it\u2019ll supplement their skills, and that all discoveries must be validated by people.<\/p>\n

\u201cThis is an exciting time to be at the intersection of artificial intelligence and the pharmaceutical industry,\u201d said BenevolentBio\u2019s Hunter. \u201cI hope to use this to get more medicines to patients faster.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"

Courtesy: NVIDIA Creating New Drugs, Faster Alzheimer\u2019s. Cancer. Parkinson\u2019s. These devastating diseases are just a few of the hundreds that scientists are struggling to cure with new medicines in the face of soaring drug-discovery costs and testing times. Now, artificial intelligence is showing the potential to be a faster, more efficient way to find and […]<\/p>\n","protected":false},"author":2,"featured_media":31516,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[14],"tags":[356,336,8,231,74],"yoast_head":"\nHow AI Promises to Speed Drug Development - Electronics Maker<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"How AI Promises to Speed Drug Development - Electronics Maker\" \/>\n<meta property=\"og:description\" content=\"Courtesy: NVIDIA Creating New Drugs, Faster Alzheimer\u2019s. Cancer. Parkinson\u2019s. These devastating diseases are just a few of the hundreds that scientists are struggling to cure with new medicines in the face of soaring drug-discovery costs and testing times. Now, artificial intelligence is showing the potential to be a faster, more efficient way to find and […]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development\" \/>\n<meta property=\"og:site_name\" content=\"Electronics Maker\" \/>\n<meta property=\"article:published_time\" content=\"2017-03-15T07:58:22+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/electronicsmaker.com\/wp-content\/uploads\/2017\/03\/Developing-a-new-medicine-is-more-expensive-and-time-consuming-than-ever.-New-methods-using-AI-could-change-that.png\" \/>\n\t<meta property=\"og:image:width\" content=\"328\" \/>\n\t<meta property=\"og:image:height\" content=\"437\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\">\n\t<meta name=\"twitter:data1\" content=\"Electronics Maker\">\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\">\n\t<meta name=\"twitter:data2\" content=\"3 minutes\">\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebSite\",\"@id\":\"https:\/\/electronicsmaker.com\/#website\",\"url\":\"https:\/\/electronicsmaker.com\/\",\"name\":\"Electronics Maker\",\"description\":\"Electronics Magazine\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":\"https:\/\/electronicsmaker.com\/?s={search_term_string}\",\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"en-US\"},{\"@type\":\"ImageObject\",\"@id\":\"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development#primaryimage\",\"inLanguage\":\"en-US\",\"url\":\"https:\/\/electronicsmaker.com\/wp-content\/uploads\/2017\/03\/Developing-a-new-medicine-is-more-expensive-and-time-consuming-than-ever.-New-methods-using-AI-could-change-that.png\",\"width\":328,\"height\":437,\"caption\":\"Developing a new medicine is more expensive and time-consuming than ever. New methods using AI could change that.\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development#webpage\",\"url\":\"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development\",\"name\":\"How AI Promises to Speed Drug Development - Electronics Maker\",\"isPartOf\":{\"@id\":\"https:\/\/electronicsmaker.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development#primaryimage\"},\"datePublished\":\"2017-03-15T07:58:22+00:00\",\"dateModified\":\"2017-03-15T07:58:22+00:00\",\"author\":{\"@id\":\"https:\/\/electronicsmaker.com\/#\/schema\/person\/df9e9cfbf590f29e49716824dd7d81f9\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/electronicsmaker.com\/how-ai-promises-to-speed-drug-development\"]}]},{\"@type\":\"Person\",\"@id\":\"https:\/\/electronicsmaker.com\/#\/schema\/person\/df9e9cfbf590f29e49716824dd7d81f9\",\"name\":\"Electronics Maker\",\"image\":{\"@type\":\"ImageObject\",\"@id\":\"https:\/\/electronicsmaker.com\/#personlogo\",\"inLanguage\":\"en-US\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/a4af77a4fcb00c5dfc7c1ca124a492b4?s=96&d=mm&r=g\",\"caption\":\"Electronics Maker\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","_links":{"self":[{"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/posts\/31514"}],"collection":[{"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/comments?post=31514"}],"version-history":[{"count":1,"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/posts\/31514\/revisions"}],"predecessor-version":[{"id":31517,"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/posts\/31514\/revisions\/31517"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/media\/31516"}],"wp:attachment":[{"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/media?parent=31514"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/categories?post=31514"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/electronicsmaker.com\/wp-json\/wp\/v2\/tags?post=31514"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}