We study the testis of male mice with a conditional knockout of syna. Knockout mice carry Pbsn-cre with floxed syna (Pbsn-cre; syna f/f), and control mice carry floxed syna without Cre (syna f/f). The genotype is the grouping we compare, and it is in the `genotype` column: "syna knockout" for the 3 knockout mice and "wild type" for the 3 syna f/f controls (the `group` column is an exact copy of `genotype`). All mice were untreated, and whole testes were collected at 4 months of age, a single time point.

Each of the six samples (KO1 to KO3 and WT1 to WT3 in `sample_id`) is the testis tissue of a different mouse, with one library per mouse, so no mouse contributed more than one sample and there are no technical replicates. The numbers only count the mice within each genotype; litters are not recorded, and there is no record that knockout and control mice were matched. The submitters recorded all six libraries in one batch (`src_batch` is 1 for every sample), so we have no batch variable, and we know of no other variable that matters; the remaining columns are GEO annotations. The GEO record says WT1 was judged an outlier and left out of the submitters' own analysis; it is still in our files and does not stand out in our own quality checks.

The data are bulk mRNA-seq (oligo-dT selected), given as raw gene-level read counts for Ensembl genes, from HISAT2 alignment to GRCm39 and StringTie quantification.

Our question is which genes are differentially expressed in the testes of syna knockout mice ("syna knockout" in `genotype`) compared with wild type syna f/f mice ("wild type"), using all six samples.