tanlabcode / limacc

LiMACC

LiMACC is an framework of calling significant interactions from multiple types of chromosome conformation capture assays, including Capture-C, Capture-HiC and HiChIP experiments. It's implemented through a R package limacc.

Install from Github

library(devtools)
install_github("wbaopaul/limacc")

Install from source codes

• Download source codes here and In R type:

install.packages('path to limacc_1.0.tar.gz', type = 'source', rep = NULL)

Note

• The latest limacc needs R version 3.5.1 or later

Required Inputs

• fragment_file A bed file (or in similar format) indicates the genomic coordinates of either the restriction enzyme fragments for Capture-C, Capture-HiC, or the binned fragments, in 5kb for instance, for HiChIP data.

  • for Capture-C or Capture-HiC data, an example of fragment_file (chromosome start end fragment_id) to indicate the genomic location of the HindIII enzyme fragment :

  chr1	0	3002504	chr1_1
  chr1	3002504	3005876	chr1_2
  chr1	3005876	3006265	chr1_3
  ......
  

  • An example of fragment_file (chromosome start end fragment_id) in 5kb resolution for HiChIP is given similarly:

  chr1	0	5000	1
  chr1	5000	10000	2
  chr1	15000	20000	3
  ......
  

  Note that the fourth column fragment_id could be any user specified character

• contact_file contact matrix file, a three column file with columns (bin_i bin_j count):

  chr1_10	chr1_11	2
  chr1_10	chr1_15	4
  chr1_15	chr1_20	5
  ......
  

  Note that the first two columns in the contact_file should be included in the fourth column of the fragment_file.

• dtype data type, 'CHiC', 'CC' or 'HiChIP'

• peakFile4HiChIP Bed file of 1D peak as virtual viewpoints for HiChIP, set as NULL when dtype is not 'HiChIP'

Output

• A data frame providing the significant interactions
• The significant interactions were also saved in a file with user specificed file name

Help in R

library(limacc)
help(limacc)

Alternative run through bash command line (for Linux and maxOS)

• copy scripts limacc4command and limacc_withArgs.R in this reporitory to the same directory, for example, current directory, then

  ./limacc4command -f your_fragment_file -c your_contact_file1,file2 -t CHiC 
  

• for more help, see

  ./limacc4command --help
  

Reference

The detailed information of LiMACC algorithm is described in the following paper: