A case study on the detailed reproducibility of a Human Cell Atlas project