Specific mutations in H5N1 mainly impact the magnitude and velocity of the host response in mice

Tchitchek, Nicolas; Eisfeld, Amie J; Tisoncik-Go, Jennifer; Josset, Laurence; Gralinski, Lisa E; Bécavin, Christophe; Tilton, Susan C; Webb-Robertson, Bobbie-Jo; Ferris, Martin T; Totura, Allison L; Li, Chengjun; Neumann, Gabriele; Metz, Thomas O; Smith, Richard D; Waters, Katrina M; Baric, Ralph; Kawaoka, Yoshihiro; Katze, Michael G

Compare to more traditional analysis methods, the Singular Value Decomposition analysis that we used for the kinetics analysis has several advantages that we exploited here. This analysis is based on an unsupervised method and does not require pulling the samples into distinct groups (i.e. the samples are not grouped by time points post-infection or by inoculation concentrations). This unsupervised aspect was crucial in our analysis because of the shifts on the host response that we described in this study. Using this SVD analysis strategy, we have been able to successfully capture the kinetics pattern of the host response system to VN1203 wild-type without having to group the samples into particular bins. Moreover, the kinetic patterns of the sub-biological mechanisms identified (eigentranscripts and eigenproteins) can be naturally displayed with this analysis strategy. These representations of eigentranscripts and eigenproteins show in a very concise way the kinetics profiles of all the associated transcripts and proteins. For example, we have been able to clearly observe the transient profile of transcripts associated with T-cell responses. With more traditional methods (i.e. gene modules, gene heatmaps or PCA representations in gene space), this kind of representation is not systematic and could produce large unintelligible representations.