# Reconstruction and analysis of the gene network regulating apoptosis in hepatocellular carcinoma based on scRNA-seq data and the ANDSystem knowledge base

## TL;DR

This study reconstructs a gene network involved in apoptosis in liver cancer using single-cell RNA data and AI methods to identify potential therapeutic targets.

## Contribution

A novel gene network for apoptosis in hepatocellular carcinoma is reconstructed using scRNA-seq and the ANDSystem knowledge base.

## Key findings

- 116 differentially expressed genes and their proteins were identified in the apoptotic process in HCC.
- Key proteins like NFKB1, BCL2, and CDKN1A showed high connectivity and differential expression in HCC.
- Proteins such as CDKN1A and ERBB2, not annotated for apoptosis, were found to interact significantly with apoptotic genes.

## Abstract

Hepatocellular Carcinoma (HCC) is the most common primary liver cancer characterized by rapid progression, high mortality rate and therapy resistance. One of the key areas in studying the molecular mechanisms of HCC development is the analysis of disturbances in apoptosis processes in hepatocytes. Throughout life apoptosis ensures the elimination of old and defective cells while the attenuation of this process serves as one of the leading factors in carcinogenesis. In this study we reconstructed and analyzed the gene network regulating hepatocyte apoptosis in humans based on single-cell transcriptome sequencing (scRNA-seq) data and the ANDSystem knowledge base which employs artificial intelligence and computational systems biology methods. Comparative analysis of gene expression revealed weakened transcription of genes involved in the regulation of inflammatory processes and apoptosis in tumor hepatocytes compared to hepatocytes of normal liver tissue. The reconstructed network included 116 differentially expressed genes annotated in Gene Ontology as genes involved in the apoptotic process (apoptotic process GO:0006915), along with their 116 corresponding protein products. It also included 16 additional proteins that, while lacking GO apoptosis annotation, were differentially expressed in HCC and interacting with genes and proteins participating in the apoptosis process. Computational analysis of the gene network identified several key protein products encoded by the genes NFKB1, MMP9, BCL2, A4, CDKN1A, CDK1, ERBB2, G3P, MCL1, FOXO1. These proteins exhibited both a high degree of connectivity with other network objects and differential expression in HCC. Of particular interest are proteins CDKN1A, ERBB2, IL8, and EGR1, which are not annotated in Gene Ontology as apoptosis participants but have a statistically significant number of interactions with genes involved in apoptosis. This indicates their role in regulating programmed cell death. The obtained results can guide the design of new experiments studying the role of apoptosis in carcinogenesis and aid in the search for novel therapeutic targets and approaches for HCC therapy using apoptosis modulation in malignant hepatocytes. Furthermore, the proposed approach to reconstructing and analyzing the apoptosis regulation gene network in hepatocellular carcinoma can be applied to analyze other tumor forms providing a systemic understanding of disturbances in key regulatory processes in oncogenesis and potential therapy targets