OnTAD is an Optimized Nested TAD caller for Hi-C data
GNU Compiler Collection (https://gcc.gnu.org/) version 7.3.1
Here are two options to use OnTAD in local:
- Users can download the executable file.
chmod +x OnTAD
Note: OnTAD is complied under gcc version 7.3.1. Please load same version to the environment, otherwise use the method below to complie the software.
- OnTAD can be complied with source code in the /src folder.
cd PathToOnTAD/src
make
chmod +x OnTAD
##Note: in some cases, static linking is neccessary. If it's needed, please add '-static' in line 3 of the makefile####
To remove OnTAD from your local folder:
make clean
Call hierarchical TADs from the test matrix (http://bx.psu.edu/~lua137/OnTAD/chr18_KR.matrix).
The test matrix is from Rao et al, Cell 2014. And it is in Gm12878 with 10kb resolution.
./OnTAD chr18_KR.matrix -penalty 0.1 -maxsz 200 -o OnTAD_KRnorm_pen0.1_max200_chr18
OnTAD result in Gm12878, chr18: 42.2Mb-44.8Mb (10Kb resolution)
Now added new optional output: genome browser compatible bed file. It provides an easy way for users to visualize their results. Each tad region is marked by a specific color corresponding to the tad level:
An example of genome browser view in mouse G1E-ER4 chr19:
*The Hi-C heatmap view is from 3D genome browser.
The OnTAD output has five columns:
startpos endpos TADlevel TADmean TADscore
Explanations of each field are as follows:
startpos & endpos: the begining and end of each TAD interval
TADlevel: the level of each TAD in the hierarchy. Small value denotes outter TADs and large value denotes inner TADs.
TADmean: average interaction frequency in each TAD region. (standardized by genomic distance)
TADscore: confidence score for each TAD.
###Note: current version requires input data in the N * N matrix format and separated by TAB or space.
OnTAD <Hi-C matrix> [-penalty <float>] [-maxsz <int>] [-minsz <int>] [-ldiff <float>] [-lsize <int>] [-bedout <chrnum> <int>] [-log2] [-o output_file]
<Hi-C matrix> the n*n Hi-C contact matrix. Both raw and normalized matrix are acceptable.
-penalty <float> The penalty applied in scoring function to select positive TADs. Higher penalty score will result in fewer TADs.
-maxsz <int> The maximum size of TADs can be called. The size is determined by number of bins covered in the contact matrix.
-minsz <int> The minimum size of TADs can be called. The size is determined by number of bins covered in the contact matrix.
-ldiff <float> The cut-off to determine local minimum. (local maximum - local minimum >= ldiff*std)
-lsize <int> The local region size that used to determine local minimum
-log2 <boolean> if specified, log2(contact frequency) will be used to call TADs.
-bedout <chrnum> <int> The chromosome number and resolution (bp), e.g -bedout chr3 10000 will generate bedfile with coordinates match chr3 under 10Kb resolution
-o <file path> The file path for the TAD calling results.
-maxsz 200
-minsz 3
-penalty 0.1
-ldiff 1.96
-lsize 5
-bedout false
version 1.1; Added option to output genome browser compatible bed format for visualization
Current version 1.2; removed requirement of GSL package
Lin An, Tao Yang, Jiahao Yang, Johannes Nuebler, Qunhua Li*, Yu Zhang*. Hierarchical Domain Structure Reveals the Divergence of Activity among TADs and Boundaries, Biorxiv 2018 (https://www.biorxiv.org/content/early/2018/07/03/361147.1)
This project is licensed under the MIT License - see the LICENSE.md file for details
This work was supported by NIH R01 GM121613 and NIH R24 DK106766 (to YZ and LA), NIH training grant T32 GM102057 (CBIOS training program to The Pennsylvania State University), a Huck Graduate Research Innovation Grant to TY, and by NIH grants R01GM109453 (to QL).
Lin An (lua137 {at} psu {dot} edu)