Science & Enterprise subscription

Follow us on Twitter

  • Our #podcast is now live ... NPC Members Display Photos At Annual Club Exhibit https://t.co/EsClVw9ZNc @pressclubdc #NPCPhotoEx
    about 2 days ago
  • More Democrats say they're in favor of increased federal spending on scientific research than Republicans, a gap th… https://t.co/SQBna2FykD
    about 2 days ago
  • New post on Science and Enterprise: Infographic – Partisan Split Remains on Research Spending https://t.co/FabPa5ufSb #Science #Business
    about 2 days ago
  • The first participant in a clinical trial received an experimental minimally-invasive brain implant to record brain… https://t.co/nPUNY6S2a5
    about 3 days ago
  • New post on Science and Enterprise: Trial Underway Testing Brain-Computer Implant https://t.co/olPDnbtfnW #Science #Business
    about 3 days ago

Please share Science & Enterprise

A.I.-Powered Network to Identify, Build Molecules

Networked earth

(geralt, Pixabay)

13 Sept. 2019. A chemical engineering team is creating an open network for chemists and data scientists to share data for speeding the development of new molecules. A pilot project for the network is funded by a nearly $1 million grant from National Science Foundation for researchers at Virginia Commonwealth University in Richmond and colleagues with other academic groups and private companies.

A team led by Tyler McQuade, professor of life science and chemical engineering at VCU is building an exchange for researchers in chemistry and engineering to contribute their data for analysis by machine learning to identify and build new molecules. The exchange, called MPrint-OKN, aims to collect quantum mechanical data on molecular performance for configuring into molecular imprints, or MPrints. The proposed open knowledge network, or OKN, offers incentives to participants to share data and benefit from more advanced mechanisms for building new molecules.

Currently, say the researchers, there are no shared repositories or portals with tools to build models of molecules or predict their performance. MPrint-OKN is expected to collect contributions from chemistry, molecular modeling, and data science to develop machine learning algorithms and data visualization tools. With more participation in the network, the tools are expected to become more refined and valuable to participants, thus encouraging wider use and more contributions.

Carol Parish, a chemistry professor at University of Richmond and a collaborator on the project, says in a joint statement, “The ability to compute molecular properties using computational techniques, and to dovetail that data with experimental measurements, will generate databases that will produce the most comprehensive results in the molecular sciences.” Parish adds, “There are many laboratories around the world working in this space, but there are few organizational structures available that encourage open sharing of this data for the benefit of the common good.”

The award for MPrint-OKN is made through NSF’s Convergence Accelerator program that aims to identify areas where research across disciplines can solve big problems with practical, near-future payoffs. Convergence Accelerator grantees are also expected to put together teams from the academic world, industry, and other sectors. The MPrint-OKN team includes co-investigators from University of Florida, as well as VCU and Richmond, and the data science companies Two Six Labs in Arlington, Virginia and Fathom Information Design in Boston. The team says several large and small companies, national labs, and National Institute of Standards and Technology are interested in taking part in MPrint-OKN.

The NSF award of $994,433 covers development through May 2020 of a pilot MPrint-OKN network that reduces the time and cost of discovering new molecules, and predicting their performance. At that point, the project team can make a pitch for funds to further develop MPrint-OKN for five years and up to $5 million.

More from Science & Enterprise:

*     *     *

Please share Science & Enterprise ...
error

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.