We studied alpha-cell-lineage cells sorted from the pancreatic islets of 5-week-old mice. The `group` column holds the genotype: Nkx2.2_alphaKO are alpha-cell-specific Nkx2.2 knockouts (Nkx2.2fl/fl;Gcg-iCre;Rosa26:tdTomato) and control_Nkx22wt are Gcg-iCre controls (Nkx2.2+/+;Gcg-iCre;Rosa26:tdTomato). From both genotypes we collected the tdTomato-positive cells by FACS. There are four control libraries (Con_1 to Con_4) and three knockout libraries, named Mut_1, Mut_2 and Mut_4; there is no Mut_3 in the files. All mice were sampled at the same age, so there is no time variable.

Each sample is one library of sorted cells, and there are no technical replicates. Whether a library comes from one mouse or from cells pooled from several mice is not recorded, but no mouse is recorded as contributing to more than one library; the numbers in the sample names only count the libraries within each genotype, and there is no record of litters or pairing. Sort, library and sequencing batches are not recorded, so we have no batch information. Sex is not recorded, and no other variable matters to us; the remaining columns are constant annotations.

The data are low-input bulk total RNA-seq (SMARTer Stranded Total RNA-Seq Pico v2). The matrix holds raw gene-level read counts from featureCounts after STAR alignment to mm10.

Our question is which genes are differentially expressed in sorted alpha-cell-lineage cells of Nkx2.2 knockout mice (Nkx2.2_alphaKO) compared with Gcg-iCre controls (control_Nkx22wt), using all seven libraries.