This training dataset is from an imaginary Staphylococcus aureus bacterium with a miniature genome. There is a reference genome in various formats as well as some fastq reads of a closely related but also imaginary mutant strain. It is a useful dataset for demonstrating: de novo genome assembly read mapping and variant calling genome annotation The files included are: wildtype.fna: the reference genome sequence of the wildtype strain in fasta format (a header line, then the nucleotide sequence of the genome.) wildtype.gff: the reference genome sequence of the wildtype strain in general feature format (a list of features - one feature per line, then the nucleotide sequence of the genome.) wildtype.gbk: the reference genome sequence in genbank format. mutant_R1.fastq and mutant_R2.fastq: Fastq sequence reads of a closely related mutant strain. The reads are paired-end. Each read is 150 bases long. The number of bases sequenced is equivalent to 19x the genome sequence of the wildtype strain. (Read coverage 19x - rather low!).