Computational Analysis of Systemic Lupus Erythematosus (SLE) using Sequencing, Systems Biology and Machine Learning Approaches

Abstract

Systemic Lupus Erythematosus (SLE), commonly known as lupus, presents a complex autoimmune challenge where the immune system mistakenly targets healthy tissues. Excessive immune responses can lead to the immune system attacking organs and tissues. This compromised immune system makes individuals more susceptible to infections, and immunosuppressive medications can heighten this risk. This study strives to unravel the intricate biological pathways associated with SLE, utilizing advanced bioinformatics and systems biology methods. The far-reaching impact of SLE on various organ systems, often leading to debilitating symptoms and reduced quality of life, highlights the urgency of this research. The major issue at hand is comprehending the underlying molecular mechanisms of SLE, with a particular focus on gene expression patterns across diverse organ systems. SLE complexity lies in its varied clinical manifestations and the variation in gene expression observed among affected individuals. Moreover, inflammatory processes driven by immune cell dysfunction, specifically the TH1 and TH2 cell pathways responsible for T-helper cell differentiation, significantly contribute to the disease progression. To address this challenge, a thorough analysis of diverse datasets was conducted, encompassing microarray, RNAseq and single cell data to examine gene expression patterns across various organ systems affected by SLE. Notably, the absence of commonly differentially expressed genes across datasets directed attention to single-cell data, valued for its ability to reveal subtle insights. Thorough examination of the expression patterns of critical genes such as Gata3, C-Maf, IFN-gamma, IL-2, IL-4, IL-5/IL-13, and IL2Ra/IL4Ra was undertaken across various concentrations using systems biology approach. These examinations yielded valuable insights into the roles of these genes in the onset and progression of SLE. Genes such as IL-2 and Gata3, were particular interest in which exhibited unusual behavior as the response increased from immune system, prompting an in-depth examination of their roles in the disease progression. Further- 2 Abstract more, a sophisticated machine learning model was developed to categorize peptides based on their affinity for binding to immune cells, such as B cells, T cells, and Major Histocompatibility Complex (MHC) molecules. Ensuring the model’s accuracy involved employing rigorous data preprocessing techniques. This model exhibited exceptional precision and accuracy (59%) in classifying peptides into three distinct groups, rigorously validated using an independent test dataset. These findings significantly enhance the understanding of the intricate molecular aspects underlying SLE. They illuminate gene expression patterns and peptide binding characteristics, establishing a robust foundation for potential therapeutic interventions and advancements in lupus diag- nostics. Future prospects encompass ongoing research to refine our understanding of SLE and develop more targeted and effective treatments, ultimately enhancing the quality of life for individuals grappling with this intricate autoimmune disorder.