Extended Data Fig. 9: CryoET data of human sperm without milling and asymmetric distribution of sperm-specific features in N-DRCs in human sperm.

a, A representative xy slice of a tomogram of human sperm (pixel size=10.648 Å, image thickness: 1, scale bar: 100 nm). Although the A- and B-tubule of the microtubule doublets are discernible, other features like axonemal dyneins and radial spokes are not clear in this slice (in contrast to Extended Data Fig. 1h-i). b, A representative xz slice of the same tomogram of human sperm (pixel size=10.648 Å, image thickness: 5, displayed in IMOD) (scale bar: 100 nm). c, A representative xz slice of a tomogram of mouse sperm lamella (pixel size=14.12 Å, image thickness: 10, displayed in IMOD). The red bar indicates how we measure the thickness of the lamella (scale bar: 100 nm). d, A representative xz slice of a tomogram of human sperm (pixel size=10.648 Å, image thickness: 100, displayed in IMOD). Gold beads on the top and bottom surfaces are circled and the red bar indicates how we measure the thickness of the ice near the cell. The top and bottom membrane of the sperm cells is not resolved due to the missing wedge so the actual thickness is larger than the measured thickness of ice in the vicinity (scale bar: 100 nm). e, A dot plot showing the thickness of lamellae of mouse sperm and ice near human sperm as a lower bound estimate of the actual cells. The means and standard deviations are indicated (for N = 69 mouse sperm tomograms and N = 59 human sperm tomograms). f, Densities corresponding to N-DRCs from the nine per-doublet averages are shown around a schematic of the (9 + 2)- axoneme. Common features among the N-DRC are colored in grey, while the unique features are highlighted in various colors.