37(A) Overview of bioinformatics pipeline. (B) Host gene expression reveals that human 38 macrophages exposed to IAV exhibit sustained production of key inflammatory mediators and 39 failure to induce expression of feedback inhibitors of inflammation. (C) Unbiased comparison with 40 total background RNA expression demonstrates that IAV cap-snatching has a strong preference 41 for, and aversion to, different groups of host transcripts. 42 IAV is an RNA virus, containing 8 negative-sense segments that are transcribed and replicated in 58 the nucleus of the host cell. As an obligate intracellular parasite, IAV is reliant on host cellular 59 machinery for replication. To accomplish mRNA production, the IAV polymerase binds directly to 60 the 5' 7-methylguanylate cap of a nascent host RNA and cleaves it roughly 10-14 nucleotides 61 downstream. The snatched sequence, known as a "leader" sequence, is employed as a primer for 62 efficient transcription of the viral mRNA (Plotch et al., 1981) and provides the cap to viral mRNA to 63 facilitate translation by host ribosomes. Previous large-scale studies of this process (Gu et al., 64 2015;Sikora et al., 2017Sikora et al., , 2014 have produced evidence that host-derived 65 RNA caps are frequently snatched from non-coding RNAs, particularly small nuclear RNAs 66 5 (snRNAs), due to their high abundance in infected cells. This has led to the conclusion that cap-67 snatching is not a selective process -that is, that host mRNAs are snatched at random (Sikora et 68 al., 2017;De Vlugt, Sikora and Pelchat, 2018). These previous RNA-Seq studies have detected 69 snatched leaders, but have been unable observe the complete pool of unsnatched sequences, 70 because of limited sequencing depth and resolution at the 5' end, both of which are necessary to 71 accurately quantify the background distribution of each host transcript. The CAGE RNA sequencing 72 method captures both host and virus-derived transcripts and, importantly, does not require a PCR 73 amplification step, thus eliminating PCR bias. 74To overcome these limitations, we utilised cap analysis of gene expression (CAGE) to sequence 75 capped RNA from the primary MDMs of 4 human donors in vitro at 4 time points over the course 76 of a 24 hour, productive infection with IAV. This allows us to observe the transcriptional response 77to IAV infection over time in unprecedented molecular detail. This work was carried out as part of 78 the FANTOM5 consortium. Data are accessible through the FANTOM5 ZENBU browser 79We employ a systems approach to identify key features of transcription during IAV infection in 82MDMs. We previously used CAGE to quantify, transcript expression, promoter and enhancer 83 activity in human MDM and produced a detailed time course profiling their response to bacterial 84 lipopolysaccharide (LPS) (Baillie et al., 2017). As in our previous work, we use the principle of 85 coexpression to identify key biological processes (Forrest et al., 2014;Baillie et al., 2016), and 86 compare the response of MDMs to bo...