hictoolsr: Hi-C analysis in R

hictoolsr is an R package with a collection of useful tools for Hi-C data analysis. The tools fall into the following broad categories:

• Extracting Hi-C Counts

  Functions that are optimized to quickly extract and aggregate data from .hic files.

• Generating Paired Interactions

  Methods to easily generate combinations of paired interactions.

• Merging/Filtering/Binning Interactions

  A collection of functions for converting to GInteractions objects and manipulating anchors.

• Plotting Functions

  Additional visualization functions that are compatible with the plotgardener package.

Installation

This package can be installed via github:

if (!requireNamespace("remotes", quietly = TRUE))
    install.packages("remotes")
remotes::install_github("EricSDavis/hictoolsr")

Workflows

We have provided a series of workflows that demonstrate how hictoolsr can be integrated and used with existing packages to analyze genomic data.

Differential Loop Calling

This workflow demonstrates how to identify and visualize differential loops between two biological conditions using the hictoolsr, DESeq2, and plotgardener packages.

Inter/Intrachromosomal Aggregate Analysis

(Coming soon) This workflow will demonstrate how to aggregate Hi-C signal from interactions between proteins bound on DNA.

Future updates

Improve function naming:

• binBedpe() args a1Pos and a2Pos
• binBedpe() to mapBedpeToBin(), mapBedpeToBin(), assignBedpeToBin(), convertBedpeToBin()?
• More general names (i.e. binAnchor() -> binGRanges())?

binBedpe()

• Improve documentation and examples

calcApa()

• Improve documentation and examples

calcBepde()

• Improve documentation and examples
• Change output from data.table to GInteractions object.
• Replace internal binning method with other binning functions

extractCounts()

• Improve documentation and examples

• Change name to extractHicCounts()

• Remove excess comments in function

• Move binning warning to its own function

• Utilize the chromosome extraction method from calcApa()

  • This might remove the chromosomes argument as well.

filterBedpe()

• Improve documentation and examples

• Currently does not filter out regions that are too close to the starts and ends of chromosomes. This is not a problem (it doesn’t affect how straw imports the data), but it might be desirable to have those interactions filtered out.

  • Specifically, remove starts - res*buffer and end + res*buffer in filtering.
  • Will require GRanges to have seqinfo

• Move binning warning to its own function

• Currently removes mcols after filtering - make sure to add those back in.

mergeBedpe()

• Improve documentation and examples.
• Change input to handle R objects (data.tables/GInteraction objects).
• Replace internal binning method with other binning functions.
• Change output from data.table to GInteractions object (or make this an argument).

plotApa()