summarise_cond_prior: Summarise a conditional prior
In andrewGhazi/malacoda: Bayesian Analysis of High-Throughput Genomic Assays

Description Usage Arguments Details Value Note Examples

This function runs Monte Carlo simulations for each variant's conditional mean priors.

1	summarise_cond_prior(cond_prior, n_cores = 1, n_samp = 50000)

`cond_prior`	a conditional prior object
`n_cores`	number of cores for parallelization
`n_samp`	number of prior simulation draws

This is used to ensure that the prior is not TOO specific. While we want the prior distribution to accurately reflect prior beliefs on a given variant's effects, it would be undesirable for the prior to pre-specify that the variant is functional. If the prior_is_func column in the output of this function is all FALSE, this means that no variant is a priori functional.

If one does get a conditional prior that has variants that are a priori functional, this can be addressed by increasing the min_neighbors argument in fit_cond_prior. See the documentation of that function for details.

Currently this function only simulates the mean parameters as in practice the dispersion parameters don't vary systematically between alleles and/or variants.

a data frame of RNA priors with prior simulation summary statistics

simulations for individual variants can be obtained with summarise_one_prior() and sample_from_prior()