Graduation Year

2021

Document Type

Thesis

Degree

M.S.P.H.

Degree Name

MS in Public Health (M.S.P.H.)

Degree Granting Department

Global Health

Major Professor

Chengqi Wang, Ph.D.

Committee Member

Derek Wildman, Ph.D.

Committee Member

Thomas Keller, Ph.D.

Committee Member

Xiaoming Liu, Ph.D.

Keywords

artemisinin resistance, chromatin accessibility, Global disease, Plasmodium falciparum, classification predictive modeling

Abstract

Malaria remains one of the immense global public health challenges, with an estimated ~200 million cases worldwide in 2019 despite the remarkable gains in reducing this deadly disease over the past decade. The recent emergence and spread of artemisinin resistance (ART-R) in Plasmodium falciparum will increasingly impede global efforts to control and eliminate malaria. Previous studies have observed broad transcriptional changes and identified several noncoding genetic variants strongly associated with ART-R. The broad transcriptional variations suggest that the malaria parasite uses sophisticated epigenetic regulation to survive under drug pressure. Therefore, evaluating the regulatory effects of noncoding-variants in malaria parasites is critical to safeguard the efficacy of frontline artemisinin-based combination therapies.In this work, we take advantage of recent advancements in Artificial Intelligence (AI) to develop a sequence-based, ab intio computational framework for rapidly predicting the regulatory effects of genetic variants. Our model directly learns a regulatory sequence code from large-scale epigenetic-profiling data, enabling prediction of epigenetic effects of sequence alterations with single-nucleotide resolution. We successfully apply this capability in recent published expression quantitative trait loci (eQTLs) and ART-R-associated variants, demonstrating the alternation of chromatin accessibility at intergenic regions linked to ART-R of the malaria parasite. We expect the deep learning model developed here to unveil the regulatory function in broad noncoding genomic regions and provide insight into crucial biological processes, like antigenic variation, gametocytogenesis, invasion, and drug resistance.

Download

Share

COinS