Skip to main content
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: Nat Neurosci. 2020 Nov 23;24(1):93–104. doi: 10.1038/s41593-020-00743-y

Extended Data Fig. 2 |. Building the in vivo intersex atlas (iViA) and registering it to fixed-brain atlas (iBNWB).

Extended Data Fig. 2 |

a, Generation of in vivo intersex atlas (IVIA): images of male (n = 5) and female (n = 8) brains expressing membranal tdTomato pan-neuronally are registered to a seed brain (reference image). Images are then transformed to generate an average model image, which, after five iterations, produces the in vivo intersex atlas (IVIA). b, IVIA registration accuracy. Left: 3D-rendering of traced tracts of Dsx+ neurons (pC2m, pC2l), and Fru+ neuron pMP7. Right: Per-axis jitter (X, Y, and Z) between matched traced tracts of pC2m, pC2l, and pMP7 across flies (n = 10, 12, and 7 brains, respectively). c, Example imaged tdTomato volumes (red scale) from different flies registered to the IVIA (right-most column, gray scale). Middle column is the intermediate registration of volumes to their own private whole brain atlas (as in Extended Data Fig. 1a). 45 out of 48 flies were successfully registered to the IVIA. d, Central brain neuropils and tracts used for point set registration to morph the in vivo intersex atlas (IVIA) to IBNWB fixed brain atlas. Points from meshes of segmented antennal lobe (AL), mushroom body (MB) (includes mushroom body lobes, peduncle and calyx), protocerebral bridge (PB), antennal mechanosensory and motor center commissure (AMMCC), anterior optic tract (AOT), great commissure (GC), wedge commissure (WEDC), posterior optic commissure (POC), lateral antennal lobe tract (lALT), posterior cerebro-cervical fascicle (pCCF), superior saddle commissure (sSADC), and whole central brain were used to generate IVIA-to-IBNWB and IBNWB-to-IVIA transformations. e, Overlay of IVIA (red) and registered IBNWB (in IVIA space, green) at different depths (90, 130, 180, 220, 260, and 280 μm). 0 μm is the most anterior section of the brain and 300 um the most posterior. f,g, Atlas-to-atlas registration accuracy measured using pC1 stalks from IBNWB and IVIA. (f) pC1 traces from IBNWB and IVIA atlases; black traces are single pC1 neurons (from IBNWB or IVIA, n = 70 and 20 pC1 traces respectively) and red trace is the mean reference pC1 stalk. (g) Between-atlas registration accuracy; IVIA-to-IBNWB transformation increases the jitter across all axes from the reference mean pC1 trace by ~2.24 μm, while IBNWB-to-IVIA transformation increases the jitter by ~2.8 μm.