new_majiq.DeltaPsi

class rna_majiq.DeltaPsi(psi1, psi2, prior, psibins=40, min_experiments_f=0.5, name1='psi1', name2='psi2')

Compute DeltaPsi between two groups of PsiCoverage (replicate assumption)

Compute DeltaPsi between two groups of PsiCoverage under assumption that the experiments in each group are replicates with the given DPsiPrior.

Parameters:

• psi1, psi2 (rna_majiq.PsiCoverage) – PsiCoverage files to be treated as replicates

• prior (rna_majiq.DPsiPrior) – Prior on deltapsi for inference

• min_experiments_f (float) – Only pass events that pass this many experiments in both groups

• name1, name2 (str) – Names to use for the aggregated groups coverage/PSI

__init__(psi1, psi2, prior, psibins=40, min_experiments_f=0.5, name1='psi1', name2='psi2')

Methods

__init__(psi1, psi2, prior[, psibins, ...])
bootstrap_logposterior([ec_idx, nchunks])	Logposterior average of deltapsi inference over each bootstrap replicate
dataset([nchunks])	Reduce to DeltaPsiDataset used for VOILA visualization
rebin(psibins)	Get DeltaPsi with different bins
smooth_logposterior([ec_idx, nchunks])	Infer deltapsi logposterior on approximation of psi bootstrapped mixture

Attributes

bootstrap_posterior	DeltaPsiPMF for average bootstrapped dpsi posteriors
discrete_logprior
discrete_prior	DeltaPsiPMF for discretized deltapsi prior
name1
name2
num_connections
passed
smooth_posterior	DeltaPsiPMF for dpsi posterior from approximated PSI posteriors