Fig. 1: Courtship is disrupted by visual threats in male flies via LC16 neurons.

a, Schematic of the courtship ritual. b, Schematic of the action selection paradigm. c, Courtship and defensive indices of wild-type males without (grey) or with (blue) the visual threat (n = 59 each). CS, Canton-S. d, LC16 split-GAL4 (Sp-GAL4) driving UAS-mCD8-GFP (green) in the adult brain; neuropil counterstaining is with anti-nC82 (magenta). Scale bar, 50 μm. e, LC16 > TNT flies fail to stop courting in response to the visual threat (n = 30, 18 and 38 (no threat); n = 32, 16 and 37 (threat)). f, LC16 > CsChrimson flies interrupt courtship upon artificial activation without threat (n = 17, 16 and 15 (red light OFF); n = 27, 21 and 20 (red light ON)). g, Schematic of live imaging with threat delivery. h,i, Left, ∆F/F₀ (%) of the LC16 > GCaMP6f signal before and after presenting a threat (h) or a non-threat light (i). Mean ∆F/F₀ (%) pre-stimulus and post-stimulus (n = 7 and 6) is also shown (right). The sample sizes represent biologically independent animals. The solid line and shaded area of live-imaging traces show mean ± s.e.m., respectively. Behavioural indexes are displayed as boxplots. The boxes delimit the lower (25th) and upper (75th) interquartile, and the horizontal line represents the median. Calcium imaging quantification plots are shown as minimum/maximum plots, and the median as a horizontal line. Each dot represents a single fly. Significant differences are indicated by different letters at the level of P < 0.05 (for example, a is different from b but not from ab). See Supplementary Table 1 for details on statistics.