Software for motif discovery and next-gen sequencing analysis

Analyzing Hi-C genome-wide interaction data

HOMER contains several programs and analysis routines to facilitate the analysis of Hi-C data.  Hi-C couples chromosome conformation capture (3C) with deep sequencing to reveal regions of genomic DNA that are in close spatial proximity in the nucleus.  Hi-C has emerged as a powerful technique to understand how the genome is packaged in cells to control gene expression.  Unlike ChIP-PET, 5C, or 4C, Hi-C is unbiased.  While HOMER can be jury-rigged to gleam information from other 3C-sequencing based methods, it has been specifically tailored Hi-C analysis.

Specialized Hi-C programs in HOMER

HOMER has several specialized programs for Hi-C analysis.  Each is covered in the tutorials below:
makeTagDirectory - special paired-end operations for making HOMER-style tag directories and filtering options for Hi-C

- primary analysis program - generates interaction matrices, normalization, identification of significant interactions, clustering of domains, generates Circos plots (most of the following programs use this one internally)

runHiCpca.pl - automated PCA analysis on Hi-C data to identify "compartments"
getHiCcorrDiff.pl - calculates the difference in correlation profiles between two Hi-C experiments
findHiCCompartments.pl - find continuous or differential regions from PCA/corrDiff results that describe what compartment regions of DNA belong to (name changed - was called getDomains.pl)

- helps automate the finding of high-resolution intra-chromosomal interactions
annotateInteractions.pl - program for re-analysis of significant interactions, such as relating them to ChIP-Seq peaks

- Novel tool to boost sensitivity by pooling features together when performing interaction calculations

3rd Party Software

The following 3rd Party [Free] Software is used by HOMER for visualizing Hi-C results.  They are required for specific functions (i.e. if you don't care about PCA, don't worry about installing R).  Most are straightforward to install:
Java Tree View - view interaction matrices (or any software to generate and view heatmaps, needed to view standard analyzeHiC output)
Circos - software to generating circle interaction diagrams. Tips on installing Circos (Needed when running analyzeHiC with "-circos" option)
R - statistical computing environment, used for PCA analysis (no special packages needed, used by runHiCpca.pl program)
Cytoscape - view processed network files created by annotateInteractions.pl and SIMA.pl

Make sure R and circos are available in your executable path as Homer will attempt to call these programs directly.  Java Tree View is strictly for visualizing output files, as is Cytoscape.

Analyzing Hi-C data with Homer

Below is a description of the general workflow of Hi-C analysis with HOMER, and each section contains detailed information about various analysis steps.

1.  Creating Tag Directories, quality control, and read filtering for Hi-C data (makeTagDirectory)
2.  Creating Background Models for Hi-C Data (analyzeHiC)
3.  Making Interaction matrices and normalizing interaction counts (analyzeHiC)
4.  Sub-nuclear compartment analysis/PCA/Clustering (runHiCpca.pl, getHiCcorrDiff.pl, findHiCCompartments.pl)
5.  Identifying significant interactions (analyzeHiC, findHiCInteractionsByChr.pl)
6.  Analysis and annotation of interactions with respect to other data types (annotateInteractions.pl)
7.  Visualizing Interactions and Other Sequencing data with Circos (analyzeHiC)
8.  Structured Interaction Matrix Analysis (SIMA.pl)

Important Tips for analyzing Hi-C data with HOMER

Can't figure something out? Questions, comments, concerns, or other feedback: