In plbaldoni/ZIMHMM: Improved Detection of Epigenomic Marks With Mixed Effects HMMs

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)
library(ZIMHMM)
library(SummarizedExperiment)

ZIMHMM

ZIMHMM (Zero Inflated Mixed effects Hidden Markov Model) is a package with a peak caller to detect broad enrichment regions from multiple ChIP-seq experimental replicates. ZIMHMM() is a multi-sample peak caller that models the zero-inflation of background counts, accounts for replicate-specific differences via a mixed effects model, and ensures that broad regions of enrichment are detected by fitting a hidden Markov model. This package also contains ZIHMM(), a fixed effects version of the peak caller. ZIHMM() can be used in single- and multi-sample settings. The function controlPeaks() provides parameters that control the Expectation-Maximization (EM) algorithm from the presented peak callers. Please refer to their documentation (e.g. ?ZIMHMM::ZIMHMM) for details. The paper for ZIMHMM was published in the Biometrics journal. Click here to read it.

Installation

You can install the released version of ZIMHMM from this repository with:

devtools::install_github("plbaldoni/ZIMHMM")

Example

The package contains an example dataset with H3K36me3 ChIP-seq (and input control) read counts from chromosome 19 of three technical replicates of Huvec cell line (ENCODE Consortium). Reads were tabulated using 500bp non-overlapping windows. An example of ZIMHMM is shown below.

# Loading example dataset
data(Huvec)

# ChIP and Control read counts, as well as the model offset.
ChIP = assay(Huvec,'ChIP')
Control = log(assay(Huvec,'Control')+1)
offset = matrix(0,nrow = nrow(ChIP),ncol = ncol(ChIP))

# Calling peaks
peakcall = ZIMHMM(ChIP = ChIP,Control = Control,offset = offset,random = 'intercept',control = controlPeaks())

Post-processing

The current implementation of ZIMHMM contains two functions for data visualization. The first function is plotPeaks, which plots the ChIP-seq read counts, the called peaks from ZIMHMM, and the associated window-based posterior probabilities of enrichment. The input data should be formatted as a RangedSummarizedExperiment object (such as the example dataset Huvec).

# Plotting peak calls
plotPeaks(output = peakcall,chr = 'chr19',ranges = c(16898093,17040483),data = Huvec)

The second function is makeBed, which outputs a BED file that can be used for visualization elsewhere (such as the UCSC Genome Browser - https://genome.ucsc.edu/).

# Exporting peak calls
makeBed(peakcall, data = Huvec,file = './HuvecH3K36me3.bed')

plbaldoni/ZIMHMM documentation built on Dec. 5, 2019, 11:43 p.m.