retrieve_pars: Retrieve model parameters

In kgori/sigfit: Flexible Bayesian inference of mutational signatures

Description

retrieve_pars obtains summary values for a set of model parameters (signatures, exposures, activities or spectrum reconstructions) from a stanfit object.

Usage

retrieve_pars(mcmc_samples, par, hpd_prob = 0.95)

Arguments

mcmc_samples	List with two elements named `data` and `results`, produced via fit_signatures, extract_signatures, or fit_extract_signatures.
par	Name of the parameter set to extract. Admits character values "signatures", "exposures", "activities" or "reconstructions".
hpd_prob	Numeric value in the interval (0, 1), indicating the desired probability content of HPD intervals (default is 0.95).

Value

A list of three matrices, which contain the values corresponding to the means of the model parameters and to the lower and upper ends of their HPD intervals, respectively.

Examples

## Not run: 
# Load example mutational catalogues
data("counts_21breast")

# Extract signatures using the EMu (Poisson) model
samples <- extract_signatures(counts_21breast, nsignatures = 2, model = "emu",
                              opportunities = "human-genome", iter = 800)

# Retrieve signatures
signatures <- retrieve_pars(samples, "signatures")

# Retrieve exposures and activities using custom HPD intervals
exposures <- retrieve_pars(samples, "exposures", hpd_prob = 0.9)
activities <- retrieve_pars(samples, "activities", hpd_prob = 0.975)

# Retrieve reconstructed catalogues (reconstructions)
reconstructions <- retrieve_pars(samples, "reconstructions")

# Plot signatures, reconstructions and mean exposures
plot_spectrum(signatures)
plot_spectrum(reconstructions)
barplot(t(exposures$mean))

## End(Not run)

kgori/sigfit documentation built on Feb. 3, 2022, 12:04 p.m.