Use of deep neural network ensembles to identify embryonic-fetal transition markers: repression of <i>COX7A1</i> in embryonic and cancer cells

Abstract

// Michael D. West 1 , Ivan Labat 1 , Hal Sternberg 1 , Dana Larocca 1 , Igor Nasonkin 2 , Karen B. Chapman 3 , Ratnesh Singh 2 , Eugene Makarev 4 , Alex Aliper 4 , Andrey Kazennov 4, 5 , Andrey Alekseenko 4, 10 , Nikolai Shuvalov 4, 5 , Evgenia Cheskidova 4, 5 , Aleksandr Alekseev 4, 5 , Artem Artemov 4 , Evgeny Putin 4, 6 , Polina Mamoshina 4 , Nikita Pryanichnikov 4 , Jacob Larocca 1 , Karen Copeland 7 , Evgeny Izumchenko 8 , Mikhail Korzinkin 4 and Alex Zhavoronkov 4, 9 1 AgeX Therapeutics, Inc., Alameda, CA, USA 2 BioTime, Inc., Alameda, CA, USA 3 Johns Hopkins University, Baltimore, MD, USA 4 Pharmaceutical Artificial Intelligence Department, Insilico Medicine, Inc., Emerging Technology Centers, Johns Hopkins University at Eastern, Baltimore, MD, USA 5 Moscow Institute of Physics and Technology, Dolgoprudny, Russia 6 Computer Technologies Lab, ITMO University, St. Petersburg, Russia 7 Boulder Statistics, Boulder, CO, USA 8 Johns Hopkins University, School of Medicine, Department of Otolaryngology-Head and Neck Cancer Research, Baltimore, MD, USA 9 The Biogerontology Research Foundation, Trevissome Park, Truro, UK 10 Innopolis University, Innoplis, Russia Correspondence to: Michael D. West, email: mwest@biotimeinc.com Alex Zhavoronkov, email: alex@insilicomedicine.com Keywords: cancer marker; Warburg effect; embryonic-fetal transition; deep neural network; stem cells Received: September 18, 2017&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Accepted: December 20, 2017&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Published: December 28, 2017 ABSTRACT Here we present the application of deep neural network (DNN) ensembles trained on transcriptomic data to identify the novel markers associated with the mammalian embryonic-fetal transition (EFT). Molecular markers of this process could provide important insights into regulatory mechanisms of normal development, epimorphic tissue regeneration and cancer. Subsequent analysis of the most significant genes behind the DNNs classifier on an independent dataset of adult-derived and human embryonic stem cell (hESC)-derived progenitor cell lines led to the identification of COX7A1 gene as a potential EFT marker. COX7A1 , encoding a cytochrome C oxidase subunit, was up-regulated in post-EFT murine and human cells including adult stem cells, but was not expressed in pre-EFT pluripotent embryonic stem cells or their in vitro -derived progeny. COX7A1 expression level was observed to be undetectable or low in multiple sarcoma and carcinoma cell lines as compared to normal controls. The knockout of the gene in mice led to a marked glycolytic shift reminiscent of the Warburg effect that occurs in cancer cells. The DNN approach facilitated the elucidation of a potentially new biomarker of cancer and pre-EFT cells, the embryo-onco phenotype, which may potentially be used as a target for controlling the embryonic-fetal transition.