Search |  Contact |  SRI Home Do not follow this link, or your host will be blocked from this site. This is a spider trap. Do not follow this link, or your host will be blocked from this site. This is a spider trap. Do not follow this link, or your host will be blocked from this site. This is a spider trap.A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A ASRI International.  333 Ravenswood Avenue.  Menlo Park, CA 94025-3493. SRI International is a nonprofit corporation.

AIC Seminar Series

The Make-It Project and Computational Synthesis of Organic Compounds

Mario LatendresseArtificial Intelligence Center[Home Page]

Date:  Thursday, January 26th 2017 at 4:00pm

Location:  EK255 (SRI E building)  (Directions)

Webex: 

Slides (SRI Wiki access required):

https://wiki.sri.com/download/attachments/4919820/Mario%20Latendresse%20-%20Make%20It%20-%2026%20January%202017.ppt?version=1&modificationDate=1489786296958&api=v2
   Abstract

The collaborative Make-It Project involves researchers from several SRI departments, including the Artificial Research Center. It is a competitive research project sponsored by DARPA (Defense Advanced Research Projects Agency) with the overall goal of designing hardware and software allowing chemists to be more productive in exploring new chemical routes to produce organic compounds and reproducing them on demand. The overall goal has three major components : a chemical flow reactor (hardware), an application to monitor it, and an application to suggest new chemical routes to produce the desired target compounds. This presentation focuses on the approach taken for the third component by the Make-It team, which we call SynRoute.

Despite many efforts, there is no well-established application that can suggest effective and workable new chemical synthesis routes. We present two of these efforts and point out their shortcomings. The approach taken in SynRoute combines 1) a large database of experimentally verified chemical reactions; 2) a small set of generic chemical transformations to create reactions that do not exist in that database; 3) an efficient algorithm to find optimal routes to produce a target compound given the large database supplemented with created reactions tailored for the target compound. A machine learning approach will be used to limit the number of created reactions, because the generic transformations, by themselves, can generate a large number of unworkable chemical reactions.

   On-line Resources