Multi-omics data integration reveals the complexity and diversity of host factors associated with influenza virus infection

Abstract

Influenza viruses pose a significant and ongoing threat to human health. Many host factors have been identified to be associated with influenza virus infection. However, there is currently a lack of an integrated resource for these host factors. This study integrated human genes and proteins associated with influenza virus infections for 14 subtypes of influenza A viruses, as well as influenza B and C viruses, and built a database named H2Flu to store and organize these genes or proteins. The database includes 28,639 differentially expressed genes (DEGs), 1,850 differentially expressed proteins, and 442 proteins with differential posttranslational modifications after influenza virus infection, as well as 3,040 human proteins that interact with influenza virus proteins and 57 human susceptibility genes. Further analysis showed that the dynamic response of human cells to virus infection, cell type and strain specificity contribute significantly to the diversity of DEGs. Additionally, large heterogeneity was also observed in protein-protein interactions between humans and different types or subtypes of influenza viruses. Overall, the study deepens our understanding of the diversity and complexity of interactions between influenza viruses and humans, and provides a valuable resource for further studies on such interactions.

Keywords: Bioinformatics; Influenza virus; Multi-omics data.

Figure 1. Data summary of genes or proteins associated with human infection of influenza viruses. For clarity, these genes or proteins were defined as Virus-Infection-associated Human Factors (VIHFs).

(A) The number of five types of VIHFs in different influenza types or subtypes. (B) The shared ratio of VIHFs between influenza types or subtypes which was colored according to the figure legend on the top left. The shared ratios of VIHFs in the upper- and lower-triangular heatmaps were calculated by taking the total number of VIHFs in the left and top type or subtypes, respectively, as the denominator, and by taking the shared number of VIHFs between the left and top types or subtypes as the nominator. (C) Overlap of VIHFs between DEGs, DEPs, DPMs and P-PPIs. (D) Overlap of VIHFs between SHFs and other kinds of VIHFs. (E) Distribution of data by viral strain, cell type and infection time points. The data above and below the black line referred to the transcriptomic and proteomic data, respectively. The blue stars indicated that data for both transcriptomic and proteomic data were available.

Figure 2. Dynamic response of human cells to influenza virus infection.

(A) The number of DEGs at different time points after infection. (B) The correlation analysis between the shared ratio of DEGs and the size of time intervals. The black line referred to the linear regression fitting and the gray area referred to the 95% confidence interval of the regressed line. (C) The shared ratio of DEGs among different time points after influenza virus infection which was colored according to the figure legend on the top right. The shared ratios of DEGs in the upper and lower triangular heatmaps were calculated by using the total number of DEGs in the left and top time points as the respective denominators, and by taking the shared number of DEGs between the left and top time points as the nominator.

Figure 3. Analysis of the strain and cell specificity in DEGs.

(A) Number of DEGs in different influenza virus strains. (B) Number of DEGs in different cell types infected by influenza viruses. (C) The shared ratio of DEGs between different strains and cell types. Asterisks (**) indicate p-value < 0.01.

Figure 4. The heterogeneity of P-PPIs in different influenza virus types or subtypes.

(A) The shared ratio of P-PPIs between different influenza virus types or subtypes which was colored according to the figure legend on the top left. The shared ratios of P-PPIs in the upper- and lower-triangular heatmaps were calculated by taking the total number of P-PPIs in the left and top type or subtypes, respectively, as the denominator, and by taking the shared number of P-PPIs between the left and top types or subtypes as the nominator. (B) The shared ratio of P-PPIs between influenza viruses by influenza virus proteins. Only the P-PPIs that interact with the virus protein were considered when calculating the shared ratio of P-PPIs between a pair of influenza virus types or subtypes.

Figure 5. The structure of H2Flu database.

Figure 6. Integration analysis of DEPs and DEGs in Calu-3 cells at 18 hpi after infection with the A/Vietnam/1203/2004 strain.

(A) Overlap between upregulated DEPs and DEGs in Calu-3 cells after infection with the A/Vietnam/1203/2004 strain. (B) Same as (A) for downregulated DEPs and DEGs. (C) GO-term functional enrichment by three categories (BP, MF, CC) and KEGG pathway analysis were performed for the concurrently downregulated DEGs and DEPs.