Figure 8. k-mer mining of BioProjects where Nnt⁺ mice were compared with Nnt^- mice.

The NCBI Sequence Read Archive (SRA) database was interrogated by k-mer mining for BioProjects where (i) some Run Accessions (listing 6J as the mouse strain) had reads compatible with a 6J background (reads for Nnt exon 2, but not exon 9) and (ii) other Run Accessions in that BioProject (listing 6J as the mouse strain) had reads not compatible with a 6J background (reads for Nnt exons 2 and 9). The methodology is described in Figure 8—figure supplement 1a, validated by BLAST alignments (Figure 8—figure supplement 1b), with raw data in Supplementary file 7d and e.

Figure 8—figure supplement 1. k-mer mining methodology and validation.

(a) Outline of bioinformatics approach. See Materials and methods for details. (b) Reads aligned to Nnt mRNA for selected accessions from Supplementary file 7. The results illustrate that where k-mer mining showed the presence of reads for exons 2 and 9, those Run Accessions had full-length Nnt mRNA (not consistent with 6J). k-mer mining showing the presence of reads for exon 2, but not 9, represent runs where exons 8–12 are missing (consistent with 6J).

Figure 8—figure supplement 2. Bruce4 ES cell line and IL28RA^-/- mouse Nnt genotypes.

(a) Integrative Genomics Viewer (IGV) Sashimi plots of RNA-Seq alignments to the Nnt gene using the C3H/HeJ reference genome. Undertaken as described for Figure 3d. Il28ra^-/- mice were available in-house and were subjected to RNA-Seq analysis. Bruce4 RNA-Seq data was obtained from SRR923434. (b) RNA-Seq analysis showed that, as expected, Il28ra (Ifnlr1) mRNA expression is lost in Il28ra^-/- mice.