Skip to content

Latest commit

 

History

History
27 lines (15 loc) · 1.24 KB

README.md

File metadata and controls

27 lines (15 loc) · 1.24 KB
Raw

License: GPL v3

mcmcneuro logo

About

The goal of MCMCneuron is to find the most representative graph model for neuronal interactions vi Markov Chain Monte Carlo.

https://brandimarte.github.io/coding/MCMC_Neuro.html

Description

graphPenalty

Computes a Markov Chain Monte Carlo on graphs representing neuronal connectivity to estimate the graph that best represents the observed data acquired from single neurons. Considers different values of the penalty constant within the 3 different following methods for computing the posterior probability:

  1. <Xi|Xj>=1 if Xi=1 and Xj=1
  2. <Xi|Xj>=1 if Xi=Xj or <Xi|Xj>=0 (if Xi!=Xj)
  3. <Xi|Xj>=1 if Xi=Xj or <Xi|Xj>=-1 (if Xi!=Xj)

The output files are sent to 'out/outPenalty' folder.

bestGraph

For a given penalty constant, chosen based on the previous executable, it computes a Markov Chain Monte Carlo on graphs representing neuronal connectivity to estimate the most representative graph for each set of observed neurons.

The output files are sent to 'out/outBestGraph' folder.