Transcriptional Profile and Machine Learning Models as Potential Tools for Non-invasive Diagnostic and Prediction of Drug Efficacy in Non-alcoholic Steatohepatitis

Summary

Nonalcoholic fatty liver disease (NAFLD) is a broad spectrum of liver disorders ranging from simple steatosis to the more progressive inflammatory form, nonalcoholic steatohepatitis (NASH). NASH incidence is growing exponentially over the years. Nowadays, 25% of the world’s population is estimated to have NAFLD with the highest rates in the Middle East (31.79%), and nearly 25% of NAFLD patients are evaluated to be NASH. However, there is no available non-invasive reliable early diagnostic tool for NAFLD/NASH diagnosis and stratification. Accumulating data have reported the role of the cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway in inflammatory liver diseases including NASH. The STING pathway is a critical signaling pathway of the innate immune system induced by cytosol DNA and it is activated by pathogen-associated molecular patterns (PAMPs) which are presented during the inflammatory events of disease progression. Integration of tissue transcriptomic data, and histological information with machine learning, may reveal new insights into NASH management.  In this study, we aim to investigate the expression of genetic-epigenetic RNA network, retrieved by in silico data analysis, related to STING signaling pathway, PAMPS, and implicated in NASH pathogenesis. As well as, applying a machine learning model to develop a novel RNA panel that could be utilized in the diagnosis and stratification of NAFLD/NASH in both human and animal model. Moreover, we aim to assess the potential therapeutic efficacy of Amloveran and DONIFOXATE on the NASH animal model via modulation of the chosen RNA network at molecular, biochemical, and histopathological levels, along with the machine learning model that will be trained to generate a score for prediction of efficacy of Amloveran and DONIFOXATE on NASH model via combined histological and gene expression variations. Ultimately, we aim to design a molecular beacon-based multiplex to detect more than one of the chosen RNA variants in one step without amplification as a simple, rapid, cheap, sensitive, and specific detection method, with a low detection limit, which will be of great value in Egypt and globally. In addition, the high impact of commercialization of these POT kits; will potentially minimize the previously used technology limitations. It is anticipated that we will likely identify the ncRNA signatures related to NASH, thus posing the basis for the possible harnessing of these ncRNAs as therapeutic targets, as well as possible diagnostic-prognostic tools. Moreover, we will adopt alternative strategies for better management of NASH with the interest in the search of new drugs and determining their mechanism of action which will be of great value in developing countries with limited resources and high incidence rates of HCC, such as Egypt. Our hope is that these minimally invasive entities provide a window to the in vivo milieu of the patients without the need for costly, complex invasive procedures, rapidly moving ncRNAs from bench to the clinic.

Achievements

List of Publications from the Project

Partners

Project Members

• Sara Hassan Abo Agawa

Project Leaders