TY - JOUR
T1 - Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules
AU - Gómez-Bombarelli, Rafael
AU - Wei, Jennifer N.
AU - Duvenaud, David
AU - Hernández-Lobato, José Miguel
AU - Sánchez-Lengeling, Benjamín
AU - Sheberla, Dennis
AU - Aguilera-Iparraguirre, Jorge
AU - Hirzel, Timothy D.
AU - Adams, Ryan P.
AU - Aspuru-Guzik, Alán
N1 - Publisher Copyright:
© 2018 American Chemical Society.
PY - 2018/2/28
Y1 - 2018/2/28
N2 - We report a method to convert discrete representations of molecules to and from a multidimensional continuous representation. This model allows us to generate new molecules for efficient exploration and optimization through open-ended spaces of chemical compounds. A deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder, and a predictor. The encoder converts the discrete representation of a molecule into a real-valued continuous vector, and the decoder converts these continuous vectors back to discrete molecular representations. The predictor estimates chemical properties from the latent continuous vector representation of the molecule. Continuous representations of molecules allow us to automatically generate novel chemical structures by performing simple operations in the latent space, such as decoding random vectors, perturbing known chemical structures, or interpolating between molecules. Continuous representations also allow the use of powerful gradient-based optimization to efficiently guide the search for optimized functional compounds. We demonstrate our method in the domain of drug-like molecules and also in a set of molecules with fewer that nine heavy atoms.
AB - We report a method to convert discrete representations of molecules to and from a multidimensional continuous representation. This model allows us to generate new molecules for efficient exploration and optimization through open-ended spaces of chemical compounds. A deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder, and a predictor. The encoder converts the discrete representation of a molecule into a real-valued continuous vector, and the decoder converts these continuous vectors back to discrete molecular representations. The predictor estimates chemical properties from the latent continuous vector representation of the molecule. Continuous representations of molecules allow us to automatically generate novel chemical structures by performing simple operations in the latent space, such as decoding random vectors, perturbing known chemical structures, or interpolating between molecules. Continuous representations also allow the use of powerful gradient-based optimization to efficiently guide the search for optimized functional compounds. We demonstrate our method in the domain of drug-like molecules and also in a set of molecules with fewer that nine heavy atoms.
UR - http://www.scopus.com/inward/record.url?scp=85042729853&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85042729853&partnerID=8YFLogxK
U2 - 10.1021/acscentsci.7b00572
DO - 10.1021/acscentsci.7b00572
M3 - Article
C2 - 29532027
AN - SCOPUS:85042729853
SN - 2374-7943
VL - 4
SP - 268
EP - 276
JO - ACS Central Science
JF - ACS Central Science
IS - 2
ER -