Fig. 6

ARTDeco analysis of donor monocytes infected with IAV reveals that eQTLs mapping to read-in genes also frequently map to upstream genes. a Distribution of median readthrough levels for top 1000 expressed genes for all samples from Quach et al. [24]. Grouped by treatment condition. b Distribution of median readthrough levels for top 1000 expressed genes for IAV samples from Quach et al. [24]. Grouped by population of origin. c Scatter plot comparing median readthrough level of top 1000 expressed genes with proportion of reads mapping to IAV NS1 gene (r² = 0.53, p < 1e-33). d Schematic of two eQTL assignments that are difficult to interpret when readthrough transcription is present. On the top, a SNP is assigned as an eQTL for both the upstream gene and the read-in gene. On the bottom, a SNP located in the upstream gene is assigned as an eQTL for the read-in gene only. The first case represents eQTLs that may modulate the expression of the read-in gene by changing the expression of the upstream gene. e Bar chart showing the number of eQTLs mapped by Quach et al. [24] to genes assigned as read-in and primary induction genes. eQTLs are classified as either mapping to the upstream gene as outlined in 6B or not mapping to the upstream gene. Enrichment was computed using Fisher exact test (p < 0.001). f Example of an eQTL (rs2661133) mapped by Quach et al. [24] that maps to both a read-in gene (SCN1B) and the upstream gene (GRAMD1A) in IAV-infected samples. Genome browser tracks corresponding to mRNA from an African Belgian (AFB) and a European Belgian (EUB) from IAV-infected and non-stimulated (NS) conditions as well as total RNA and H3K27ac for IAV infection from Heinz et al. [9]. The readthrough region is outlined in the black box