FlyBrainLab
Interactive platform for exploring fly brain data, constructing executable circuits, and studying their modeled function.
Neurons are simulated — leaky integrate-and-fire or better — with documented, tuned parameters.
Dispute this rating →The evidence
No evidence, no level.The platform’s neuron models are real and documented: Neuroballad ships LeakyIAF with resting and reset potential, threshold, capacitance and resistance, alongside Hodgkin-Huxley and Connor-Stevens models with gating states.
https://github.com/FlyBrainLab/Neuroballad/blob/b931c41578f2ec3118a52d293bfa45230cfe49e9/neuroballad/models/neurons.py#L3-L20An interactive platform for exploring fly brain data, constructing executable circuits, and studying their modeled function.
https://github.com/FlyBrainLab/FlyBrainLab
L1 for executable circuits with documented neuron dynamics — but the evidence took a second reading to place correctly. The FlyBrainLab repository itself is thirteen files of installers, Dockerfiles and a README; the neuron models live in the same organisation’s Neuroballad, which is what the level rests on. Not L2: the platform can host plasticity, and hosting a rule is not running one — L2 would need a specific modelled circuit shipped with it in which synapses change. No validation mark: the eLife 2021 paper uses physiological recordings as model inputs (odorant binding affinities), which is calibration, and its one in-vivo reference is a sentence noting that activity bumps persisted as previously observed, with no figure putting recorded fly data beside model output. A qualitative citation that a model behaves the way an experiment reported is below the bar.
Record
Nearby
All projects →Ratings on thelearningfly.com are proposals, not verdicts. Every one of them is arguable in public.