Home> News>Detail

In the future, artificial intelligence will invent new drugs sooner than humans

    Keyword:
Time:20170525

Today, artificial intelligence algorithms can perform very detailed data analysis through depth learning, from facial recognition to medical impact analysis, and the performance of artificial intelligence algorithms has caught up and even surpassed human performance. More and more high-tech applications are being used by major pharmaceutical companies in the field of new drug research and development, and they hope to explore ways to improve the efficiency of new drug research and save more costs.

.

Atomwise, a start-up company in San Francisco and Y Combinator company, has established a AtomNet (PDF) system, the company to potential Ebola virus and multiple sclerosis disease drug production. The project aims to simulate the pharmaceutical process using supercomputers, AI and complex algorithms to predict the effectiveness of new drugs while reducing R & D costs. Atomwise has launched two programs that show the potential of AtomNet, multiple sclerosis drugs and Ebola virus. According to Atomwise, MS drugs have been licensed to UK pharmaceutical companies that are not disclosed, and Ebola drugs are being prepared for submission to peer-reviewed publications.

Atomwise recently used AI technology to analyze and test more than 7000 drugs in less than a day, contributing to the search for a cure for ebola. According to the company's statistics, if the use of traditional methods, this analysis will take months or even years to complete. But Alexander Levy, chief operating officer at Atomwise, mentioned that AtomNet also needs to be tested and that artificial intelligence does not solve all medical development problems. In addition to the discovery of new drugs, the Berg biopharmaceutical company developed new drugs by studying biological data. "Berg through its development of Interrogative Biology artificial intelligence platform, research on human health research organization, human molecular and cellular defense organization and pathogenesis principle mechanism, to calculate the body's own molecular potential drug compounds based on artificial intelligence and big data.

Due to the cost of trial and error is too high, more and more manufacturers began to turn to the drug development of computer and artificial intelligence, hope to narrow the range of potential drug molecules by using this technology, which can save time and money in subsequent tests. To identify the coding genes for proteins that have great potential to serve as drug targets, these vendors place their hopes on algorithms. At present, some new algorithm model increases the complexity of the new level, used to narrow the proteins, drugs and clinical data, in order to better predict which genes are most likely to get protein and drug combination. Researchers estimate that about 15%~20% of new drug costs are spent in the exploratory phase. Normally, this means spending up to hundreds of millions of dollars, as well as 3~6 years of work. Today, it is hoped that the process will be shortened through AI to a few months, and significantly reduce R & D costs.

TwoXAR is developing a AI driven glaucoma drug, and Berg is working on an algorithm for cancer treatment. Atomwise's project is unique in that it extracts large amounts of data from life to death. Because it involves a lot of expensive and time consuming drugs, this function solves the "life and death" problem in the pharmaceutical industry. Atomwise says it has the world's best results in new drug discovery, binding, affinity prediction, and toxicity testing. In terms of partners, Atomwise has continued to work with academic and business customers in addition to some confidential projects with Merck and Autodesk.

Artificial intelligence, the development of new drugs enterprises gradually increased, in addition to European and American pharmaceutical companies, Japanese pharmaceutical companies are also actively facing new technologies. We believe that, as researchers find new patterns in the existing knowledge system, there will be a wave of medical innovation that will not be overlooked.