Set of functions and scripts to process the data used in the submission: "Distributed context-dependent choice information in mouse posterior cortex"

• MATLAB R2020 or after
• For locaNMF (based on pyTorch). See: https://github.com/ikinsella/locaNMF
• For the RNN (standard tensorflow2 installation with GPU support)

• main.m is used to prepare already preprocessed imaging and behavioral data from Attention separates sensory and motor signals in the mouse visual cortex
• runLocaNMF.ipynb - Python notebook to run locaNMF (after main.m)
• locaNMFplots.m - Produce output plots from locaNMF statistics
• computeStateVectors.m - main state vector computation used in the manuscript
• plotStateVectors.m - several plots related to the state vectors
• RNNmodel.py - generates synthetic data following the animals psychometric responses and trains the RNNs used in the manuscript
• figures folder contains .mat files and .m scripts to generate most of the figures presented in the paper