Investigation of HIV-TB co-infection through analysis of the potential impact of host genetic variation on host-pathogen protein interactions

HIV and Mycobacterium tuberculosis (Mtb) co-infection causes treatment and diagnostic difficulties, which places a major burden on health care systems in settings with high prevalence of both infectious diseases, such as South Africa. Human genetic variation adds further complexity, with variants affecting disease susceptibility and response to treatment. The identification of variants in African populations is affected by reference mapping bias, especially in complex regions like the Major Histocompatibility Complex (MHC), which plays an important role in the immune response to HIV and Mtb infection. We used a graph-based approach to identify novel variants in the MHC region within African samples without mapping to the canonical reference genome. We generated a host-pathogen functional interaction network made up of inter- and intraspecies protein interactions, gene expression during co-infection, drug-target interactions, and human genetic variation. Differential expression and network centrality properties were used to prioritise proteins that may be important in co-infection. Using the interaction network we identified 28 human proteins that interact with both pathogens (”bridge” proteins). Network analysis showed that while MHC proteins did not have significantly higher centrality measures than non-MHC proteins, bridge proteins had significantly shorter distance to MHC proteins. Proteins that were significantly differentially expressed during co-infection or contained variants clinically-associated with HIV or TB also had significantly stronger network properties. Finally, we identified common and consequential variants within prioritised proteins that may be clinically-associated with HIV and TB. The integrated network was extensively annotated and stored in a graph database that enables rapid and high throughput prioritisation of sets of genes or variants, facilitates detailed investigations and allows network-based visualisation.

Grantor

Department of Clinical Laboratory Sciences

Year dc:date.issued

2022

Author dc:creator

• Heekes, Alexa Storme

Advisor dc:contributor.advisor

• Mulder, Nicola

• Bioinformatics

Handle dc:identifier.uri

http://hdl.handle.net/11427/36779

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/36779