Fig. 1: Spatial clustering of ocular dominance in layer 4 of mouse visual cortex.
From: A column-like organization for ocular dominance in mouse visual cortex
a Multiplane, large-field of view (FOV) two-photon calcium imaging of ocular dominance in layer 4 of the visual cortex. Left: Schematic indicating volume dimensions and acquisition rate. Middle: Example FOV (GCaMP6s expression) of a single plane; “A”, “P”, “M“, and “L” indicate anterior, posterior, lateral, and medial. Right: HLS map for ocular dominance. Hue: Eye-preference (contralateral: Blue; ipsilateral: Red); Lightness: ΔF/F; Saturation: Eye selectivity. Scale bars: 100 μm. For maps of all nine mice, see Supplementary Fig. 12c. b Left: ΔF/F response to drifting grating stimuli for three example cells (see a). “C”: Contralateral, “I”: Ipsilateral. Scale bar, vertical: 100% ΔF/F, horizontal: 10 s. Right: Inferred spiking activity of all visually responsive cells (n = 1834) in (a), sorted vertically by ODI and aligned horizontally to the preferred (P) direction (N: Null direction; blue/red: Contra/ipsi). Scale bar, vertical: 100 neurons, horizontal: 10 s. c Left: Local density of ipsilateral (top) and contralateral (bottom) eye preferring visually responsive neurons separately (across four-plane volume). Black line: lateral boundary of V1 (V1 is to the right). Black circles: Ipsi-cluster centers (see “Methods”). Right: All visually responsive neurons, color coded for ODI. White iso-ODI lines indicate ODI = 0 (solid) and ODI = 0.2 (dashed). Scale bar: 100 μm. d Histogram of ODI (for volume in c). e Mean ODI as a function of distance to the three ipsi-cluster centers in (c) (individual ipsi-clusters in gray). Red line: sigmoid fit; the point of maximum inclination (red arrow) approximates the ipsi-cluster radius. f Preferred azimuth (top) and elevation (bottom) of the neurons shown in (c). Scale bar: 100 μm. a–f Data of mouse M02. g Left: ODI as function of distance to ipsi-cluster centers. Black: Mean ± s.e.m., gray: Individual mice (n = 9). Right: Mean ODI inside (“In”, 0–100 μm) and outside (“Out”, 100–200 μm) ipsi-clusters (two-sided WMPSR test, W = 0, p = 0.004, n = 9 mice). h Same as (g) black line (mean ± s.e.m.) shows actual data (“D”), blue, green, and red lines show global randomization controls. Right: Mean (±s.e.m.) ODI “In” ipsi-clusters for real and shuffled data (two-sided Kruskal-Wallis test, H3 = 17.7, p = 5.0 × 10−4, post hoc two-sided WMPSR test, **p < 0.01, n = 9 mice). Blue, “S”: ODI values shuffled across neurons. Green, “U”: XY coordinates of neurons resampled from uniform distribution. Red, “R”: Ipsi-cluster centers randomly placed in FOV. i As (h), colored lines show local randomization controls in which the positions of pairs of neurons, spaced at a distance of 50 μm (dark red) to 250 μm (yellow), were swapped (two-sided Kruskal-Wallis test, H5 = 16.8, p = 0.0048, post hoc two-sided WMPSR test, ns not significant, **p < 0.01, n = 9 mice).
