Skip to main content
Figure 5 | BMC Genomics

Figure 5

From: Dual 3’Seq using deepSuperSAGE uncovers transcriptomes of interacting Salmonella enterica Typhimurium and human host cells

Figure 5

Confirmation of time-dependent expression patterns from pathogen as well as host cells and evaluation of dual 3’Seq quantification accuracy by quantitative real-time PCR. (a) Biological variance of the targeted transcripts across three independent biological replicates. Means with SD of the ∆Ct values from non-interacting cells (green) and cells from early (vermillion), mid-level (orange) and late interaction (blue) are shown for 19 mRNAs involved in host cell responsaes to bacterial infection (upper graph) as well as 24 prokaryotic transcripts (lower graph). Additional coding sequences of polycistronic transcripts from the prokaryote are indicated by brackets. ∆Ct values are plotted on an inverted ordinate, where a low figure reflects an increased abundance of the respective transcript and vice versa. For each graph, the lowest Ct value was arbitrarily shifted to zero followed by an according normalization of the other values. (b) Heat map of time-dependent log2 fold changes and corresponding ∆∆Ct values from early and mid-level interacting as well as mid-level and late interacting host cells. The individual expression ratios determined by qRT-PCR from three biological replicates (qPCR) together with the corresponding ratios identified by deepSuperSAGE (NGS) are depicted after hierarchical clustering of all samples based on Uncentered Pearson Correlation. Upregulated transcripts are represented in red, while downregulated transcripts are shown in green. (c) Principal component analysis of the data included in the heat map along with the corresponding log2 fold changes from non-interacting and early interacting cells (green). (d) Scatterplot depicting the relative quantification accuracy of dual 3’Seq in comparison to probe-based qRT-PCR. The expression ratios of 23 polyadenylated mRNAs and their non-polyadenylated degradation intermediates in early interacting cells determined either by deepSuperSAGE (SSage) or MACE (y-axis) is plotted against the respective qRT-PCR-based ratios (x-axis). An optimal correlation is indicated by the diagonal line.

Back to article page