INFERRING GENE REGULATORY NETWORK FOR THE SPOROGENIC STAGE OF PLASMODIUM FALCIPARUM LIFE CYCLE USING GRAPH NEURAL NETWORKS AND scRNA-SEQ DATASET

ONYEMAECHI, Promise and Covenant University, Theses Masters (2024) INFERRING GENE REGULATORY NETWORK FOR THE SPOROGENIC STAGE OF PLASMODIUM FALCIPARUM LIFE CYCLE USING GRAPH NEURAL NETWORKS AND scRNA-SEQ DATASET. Masters thesis, Covenant University.

PDF Download (142kB)

Abstract

Malaria, primarily caused by the parasite Plasmodium falciparum, remains one of the most severe infectious diseases globally, particularly in sub-Saharan Africa, where it leads to significant morbidity and mortality. A deep understanding of the molecular mechanisms that govern P. falciparum infection, especially during the sporogonic stage of the parasite’s life cycle, is crucial for controlling malaria and identifying new therapeutic targets. In this context, gene regulatory networks (GRNs) provide a valuable framework for exploring the gene interactions that drive the parasite's life cycle and its response to environmental changes. This research aims to the GRNs for the sporogenic stage of P. Falciparum life cycle using Graph Neural Networks (GNNs), a deep learning technique applied to single-cell RNA sequencing (scRNA-seq) data. With the advent of scRNA-seq, gene expression can now be studied at the individual cell level. This is particularly important in the life cycle of P. falciparum, whose sporogonic stage is heterogeneous and involves different cell types throughout its life cycle. The study began with the preprocessing of scRNA-seq data, followed by clustering the cells to identify distinct stages of the parasite. A correlation-based co-expression network was then constructed to capture the relationships between genes within these clusters. Graph convolutional neural networks were employed to reconstruct the GRNs, leveraging their ability to learn the interactions between genes in the network. The performance of the inferred GRNs was evaluated using the AUC, AUPRC ratio, and EPR metrics, with average values of 0.6756, 1.51, and 2.48, respectively. This research not only enhances the understanding of the sporogonic stage of the life cycle of P. falciparum's gene regulatory mechanisms but also identifies potential master regulators and genes that play central roles in the parasite's survival and pathogenicity. Four master regulators were identified for four selected clusters of genes associated with the life cycle of the sporogonic stage, based on the centrality scores of genes in the network. Among these master regulators, PF3D7_0304500, PF3D7_1357200, PF3D7_0822300, and PF3D7_0212300 stood out as the most critical genes within their respective clusters.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Plasmodium falciparum, malaria, Graph Neural Network, scRNA seq, Gene regulatory network, master regulator
Subjects:	Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QR Microbiology
Divisions:	Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User:	Patricia Nwokealisi
Date Deposited:	30 Sep 2024 13:05
Last Modified:	30 Sep 2024 13:05
URI:	http://eprints.covenantuniversity.edu.ng/id/eprint/18472

Actions (login required)

View Item