Figure 2

Annotation statistics for the generated expression matrices and differentially expressed genes of interacting pathogen and host cells. (a,b) Read numbers annotated to human (green) and pathogen (brown) cells or both (blue) for all sequenced libraries, respectively. The bar graph shows log₁₀-transformed numbers, while the original number of reads is depicted in a circle graph. (c) Classification of unambiguously annotated transcripts from host cells into mRNAs (vermillion), non-coding RNAs (yellow), bivalent transcripts encoding mRNAs and non-coding RNAs (orange), long non-coding RNAs (purple), histone-encoding mRNAs (reddish purple), and other transcript classes (black) for the poly(A)⁺ and poly(A)⁻ transcriptome, respectively. mRNAs encoding cytoplasmic (skyblue) or mitochondrial (blue) ribosomal proteins are depicted individually for the poly(A)⁺ and combined (bluish green) for the poly(A)⁻ transcriptome that additionally comprises small nucleolar RNAs (skyblue) and small nuclear RNAs (blue). Respective abundances are shown for non-interacting (circle graphs) as well as interacting (radial graphs) cells. The different interaction stages (early, mid-level, and late interaction) prepared with 3’Seq coupled to deepSuperSAGE (SSage) are depicted from inside outwards along with the early interaction library that was prepared using MACE (doubled width). (d) Venn diagram showing the number of commonly expressed genes for all combinations of libraries corresponding to one of the expression matrices. Ambiguously annotated reads are not included. (e) Pair-wise comparison of differentially expressed transcripts within the expression matrices (black) along with the number of exclusively expressed transcripts (gray) for each library. Listed are unambiguously annotated reads with a log₂ fold change stronger than |1.5| and an FDR-corrected p-value below 0.05. Up- and downregulated transcripts are indicated by corresponding arrows for the library on the ordinate, respectively.