Kadurin A(1,2,3), Nikolenko S(4,2,3), Khrabrov K(5), Aliper A(1), Zhavoronkov A(1,6,7).
(1) Pharmaceutical Artificial Intelligence Department, Insilico Medicine, Inc., Emerging Technology Centers, Johns Hopkins University at Eastern , Baltimore, Maryland 21218, United States.
(2) Steklov Mathematical Institute at St. Petersburg , St. Petersburg 191023, Russia.
(3) Kazan Federal University , Kazan, Republic of Tatarstan 420008, Russia.
(4) National Research University Higher School of Economics , St. Petersburg 190008, Russia.
(5) Search Department, Mail.Ru Group Ltd. , Moscow 125167, Russia.
(6) The Biogerontology Research Foundation , Trevissome Park, Truro TR4 8UN, U.K.
(7) Moscow Institute of Physics and Technology , Dolgoprudny 141701, Russia.
Deep generative adversarial networks (GANs) are the emerging technology in drug discovery and biomarker development. In our recent work, we demonstrated a proof-of-concept of implementing deep generative adversarial autoencoder (AAE) to identify new molecular fingerprints with predefined anticancer properties. Another popular generative model is the variational autoencoder (VAE), which is based on deep neural architectures. In this work, we developed an advanced AAE model for molecular feature extraction problems, and demonstrated its advantages compared to VAE in terms of (a) adjustability in generating molecular fingerprints; (b) capacity of processing very large molecular data sets; and (c) efficiency in unsupervised pretraining for regression model. Our results suggest that the proposed AAE model significantly enhances the capacity and efficiency of development of the new molecules with specific anticancer properties using the deep generative models.