mu.recal

a vetor of the same length as number of samples: the non-reference counts of n samples at the ith position.

a vetor of the same length as number of samples: the coverage of n samples at the ith position.

a vetor of the same length as number of samples: posterior probability of RR of n samples at the ith position.

a vetor of the same length as number of samples: posterior probability of RV of n samples at the ith position.

a vetor of the same length as number of samples: posterior probability of VV of n samples at the ith position.

a vetor of the same length as number of samples: the estimator of delta(sample effects).

delta

a single value representing the estimator of eta.

mu.recal calculates the optimal estimator of the ith position effect when given other parameters.

Genotyping the population using next generation sequencing data is essentially important for the rare variant detection. In order to distinguish the genomic structural variation from sequencing error, we propose a statistical model which involves the genotype effect through a latent variable to depict the distribution of non-reference allele frequency data among different samples and different genome loci, while decomposing the sequencing error into sample effect and positional effect. An ECM algorithm is implemented to estimate the model parameters, and then the genotypes and SNPs are inferred based on the empirical Bayes method.

mu.recal: One step maximization of mu when given other parameters

Description

Usage

Arguments

Value

References