GitHub - jrreimer/Long-transient-dynamics-in-the-presence-of-noise: R scripts and supporting files used in manuscript "Detection of long transient dynamics in noisy time series".

jrreimer / Long-transient-dynamics-in-the-presence-of-noise Public

Notifications You must be signed in to change notification settings
Fork 0
Star 1

R scripts and supporting files used in manuscript "Detection of long transient dynamics in noisy time series".

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
data		data
fig		fig
Figure 1.Rmd		Figure 1.Rmd
Figure 2.Rmd		Figure 2.Rmd
Figure 3.Rmd		Figure 3.Rmd
Figure 4.Rmd		Figure 4.Rmd
README.txt		README.txt
basis_functions_plot.R		basis_functions_plot.R
kl_combinations.R		kl_combinations.R
kl_divergence.R		kl_divergence.R
longTransients_1.1.tar.gz		longTransients_1.1.tar.gz
neg_grad_potential_plot.R		neg_grad_potential_plot.R
replicates.csv		replicates.csv
simulate_trajectories.R		simulate_trajectories.R
simulation_study.R		simulation_study.R
single.csv		single.csv
single_combinations.R		single_combinations.R

Repository files navigation

Simulation study for Bayesian hierarchical model.

R package longTransients must be installed:

  install.packages("path/to/longTransients_1.1.tar.gz", repos = NULL)

simulate_trajectories.R creates the simulated data stored in /data

Scripts Figure 1.R, Figure 2.R, Figure 3.R and Figure 4.R create corresponding Figures 1-4 in the manuscript. 

kl_combinations.R fits the parametric model to subsets of the simulated trajectories used to create Figure 5.

kl_divergence.R summarizes the results of the fits in kl_combinations.R through numerical approximations of marginal KL divergences to create Figure 5.

neg_grad_potential_plot.R creates Figure 6.

The simulation_study.R script fits both models to several subsets of the 10 simulated trajectories for a single value of the measurement error variance. The script also creates Figure 7.

single_combinations.R fits both models to one subset of the simulated trajectories.