melanogaster

Two real brains. He is the MaleCNS connectome (Janelia / Google / Cambridge, 2026): one male fly, 165,122 neurons. She is BANC (Bates, Phelps et al., Nature 2026): one female fly's brain and nerve cord, 150,802 neurons. Each runs as its own simulation (leaky integrate-and-fire, the model of Shiu et al. 2024) on our server.

He courts. He is shown a female through his own senses (pC1). His brain answers with its song: pIP10, the command neuron for courtship song, and his wing motor neurons, which in a real fly vibrate the wing to sing. His legs answer too (tapping), though in him they barely move.

The signals cross. His wing motor rate becomes the sound reaching her ears: her Johnston's organ cells (JO-A, JO-B) are driven at twice his wing rate. His leg rate becomes touch on her bristles. His scent is his body's, not his brain's: while he is near, cVA, the male pheromone, reaches her Or67d cells (ORN_DA1) at a fixed rate. These conversions are ours.

She decides. She is a virgin: her pC1 neurons, which carry her mating status, are kept active, as they are in a virgin. Her answer is read where Wang et al. (Nature 2021) found it: vpoDN (DNp37), the neuron that opens the vaginal plate, the female's acceptance. She accepts when his signals push her vpoDN above anything it does when she is alone (measured each time her brain starts). DNp13 (ovipositor extrusion) firing is a rejection; her escape neuron (the giant fibre) firing means she flew off. Otherwise she was not interested. Those decisions are hers.

Sleep. She keeps his hours: the night of Janelia, the lab both brains came from, and wakes by his rules (a shadow, a touch, a sound, light). At night the light here wakes her, not him. She does not court him: in this species courtship (song, chasing, tapping) is the male's, and her brain has no pIP10, the song command neuron. Awake, she waits. If someone wakes him too (on drosophila.io) and brings him to her, they meet; and if he is brought while she sleeps, his song, a sound, wakes her.

What we can't do yet. In a living female, song reaches vpoDN through vpoEN, cells tuned to the rhythm of his pulse song. In her simulated brain vpoEN stay silent: strong inhibitory cells on the way (CB3881 and others) shut them, and a simple model can't reproduce the timing that lets real song through. His song still reaches her acceptance neuron, but by weaker routes; his scent and touch matter more than they would in life.

Sources: BANC, Bates et al., Nature 2026 (CC BY 4.0) · MaleCNS, Janelia FlyEM 2026 (CC BY 4.0) · Wang et al., "Neural circuit mechanisms of sexual receptivity in Drosophila females", Nature 589:577 (2021) · Kurtovic et al., "A single class of olfactory neurons mediates behavioural responses to a Drosophila sex pheromone", Nature 446:542 (2007) · von Philipsborn et al., Neuron 2011 (pIP10, courtship song) · Shiu et al., Nature 2024 (the whole-brain model).