abc-ribbon: Approximate Bayesian Inference and a Stochastic Model of a Ribbon Synapse
Code accompanying the paper "Approximate Bayesian Inference for a Mechanistic model of Vesicle Release at a Ribbon Synapse" (Schröder, James et al. Preprint: https://doi.org/10.1101/669218 , NeurIPS version: https://papers.nips.cc/paper/8929-approximate-bayesian-inference-for-a-mechanistic-model-of-vesicle-release-at-a-ribbon-synapse )
data - folder:
Contains the stimulus and experimental recordings for the release of two BC in preprocessed version. Also the dF/F example traces for one cell are stored. And a notebook for showing the data.
generalized_method - folder:
Contains a simple example how to generalize the presented method to other problems. Including the essential steps of defining a meaningful loss function and prior distributions. The essential sampling and updating functions are in abc_method.py and not specific to the model.
poster - folder:
Contains the poster which will be presented at NeurIPS conference.
paper_version - folder:
Contains all files for the presented ABC method. Contains all files to reproduce the plots of the paper.
If you are only interested in the stochastic model of the ribbon synapse, look here.
ribbonv2.py: contains the ribbon model
standalone_model.ipynb: running the model in ribbonv2.py with specified stimulus and model parameters.