Whole-blood transcriptional signatures of different severity profiles in patients with influenza A H1N1 infection Open Access

Gu, Chunhui (Spring 2020)

Permanent URL: https://etd.library.emory.edu/concern/etds/kw52j9101?locale=en


Background: As influenza remains one of the major threats worldwide because of its high mutational and recombination rates, vaccines and antivirals continue to be developed that help the body fight against infection through adjusting host gene expression for as many strains of the virus as possible.

Objective: The primary purpose of this study is to investigate which genes or gene sets are related to the development of a high severity profile in patients with influenza infection. 

Methods: Normalized microarray data were obtained from a prior publicly available influenza study (Dunning et al. Nat Immunol, 2018). Differential expression analysis was accomplished by R version 3.6.1 and “limma” package version 3.40.6. Gene set enrichment analysis (GSEA) was conducted by using Molecular Signatures Database (MsigDB). CibersortX, an enhanced digital dissection technique was used to generate cell-specific gene expression patterns. 

Results: The differential expression analysis (DEA) showed that 48 (236), 158 (323), and 293 (425) genes were differentially down-regulated (up-regulated) in severity I, severity II, and severity III categories of influenza patients, respectively, as compared to healthy controls (the changes of expression in severity categories compared with healthy controls were called contrasts for short). In total, there were 824 distinct differentially expressed genes (DEGs). 138 of the 824 DEGs were considered to have obvious different fold-changes under different contrasts and the difference in fold-change mainly came from the comparison between high severity contrast and low severity contrast. Several gene sets were identified based on gene set enrichment analysis and over-representation analysis. 

Discussion: The analysis of overlaps between enriched gene sets of different severity profiles of patients with influenza (compared to healthy controls) demonstrated that autoimmune mechanisms could play an important role in the development of a high severity profile. There remained around half of 138 DEGs that could not be attributed to any gene set in target collections. A custom set can be built from those genes as possible gene set describing the development of influenza A pathogenesis. The cell-specific transcriptional signature generated by CibersortX suggested that neutrophils and monocytes represent the main cell-types in which the effect of those genes and gene sets takes place.

Table of Contents

Introduction 1

Methods 4

Results 7

Discussion 11

References 15

Figures and tables 17

Table 1. Descriptive table of samples grouped by severity 17

Figure 1. Gene-level analysis 19

Figure 2. Gene set enrichment analysis 21

Figure 3. cell-type-specific analysis 22

Appendix 1: Supplementary figure and table 23

Appendix 2: Supplementary methods 31

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files