Viral fitness predicts the magnitude and direction of perturbations in the infected host transcriptome

Héctor Cervera; Silvia Ambrós; Guillermo P. Bernet; Guillermo Rodrigo; Santiago F. Elena

Thirdly, ANOVA tests were performed to identify genes that vary across all conditions (with FDR as above, adjusted P < 0.05). To identify the genes shown in Fig. 3A , the test was done over all samples, including the control. By contrast, to identify the genes shown in Fig. 6A , the test was done over all samples corresponding to infections with distinct TEV genotypes. An additional criterion of significant Spearman's correlation between mean fitness and mean expression (P < 0.05) was imposed in this latter case. Lists of genes whose expressions correlate with viral fitness, either positive or negative, provided in Supplementary Dataset S4. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/206789 doi: bioRxiv preprint 20 provided by Agilent) was updated by BLASTing the oligo sequence file (021113_D_Fasta_20130122.txt) against the most recent version of the N. tabacum mRNA database (Ntab-BX_AWOK-SS_Basma.mrna.annot.fna) available at the Sol Genomics Network 77 (SGN). Sequences not returning a significant BLAST hit were removed from the output. A total of 40,430 annotated probes were generated. In 2,673 cases, more than one probe pointed to the same N. tabacum gene (e.g., probes A_95_P217927 and A_95_P046006 were both complementary to gene EB438730), and in those cases the target appeared twice in the output.