Comparison tool for Topologically Associated Domains
TADs are smaller structural units of chromosomes that are defined as regions whose DNA sequences preferentially contact each other.
The three-dimensional (3D) genome organization has the importance of tin regulating gene expression and other genomic processes. Many such domains change during development or disease, and exhibit cell- and condition specific differences.
In Hi-C maps, TADs are represented as blocks along the diagonal with sizes ranging from about 100 kilobases to 2 megabases, and they indicate increased interactions among chromatin elements within the domain compared to the upstream and downstream regions.
We developing CTADo for differential analysis of boundaries of interacting domains between two Hi-C datasets. Our tool aimed at providing a accurate, user-friendly approach to differential analysis of boundaries of TADs.
We introduce a method based on the given TAD boundaries and use it to identify four types of boundary changes: intensity change, shifted, merge and split.
In summary, CTADo is providing analyses from boundary calling to visualization.
CTADO documentation is available on the GitHub wiki page.
To get the tool CTADo clone the git repository:
git clone https://github.com/ivandkoz/CTADo.git && cd CTADo
To create conda enviroment using enviroment.yml file:
conda env create --name ctado -f environment.yml && conda activate ctado
To run the CTADO.py script, you can call it from the directory where the tool is located.
For tool usage the names of mcool or cool files, resolution, window, binsize and file that contains TADs insulating boundaries must be provided.
This file can be retrieved using:
- cooltools
- Armatus
- use the built-in function (based on cooltools TADs boundary markup) - you can run script without providing TADs insulating boundaries file name (exanple in GitHub wiki)
Vizualisation the results of intensity type of boundary changes are included in CTADo.py.
By default, the program creates the top 5 most changed by intensity type of TADs. To use it, mcool or cool files, resolution, window, binsize, both files that contains TADs insulating boundaries, and result of CTADO.py script - dataframe with intensity information - must be provided.
You can see an example of graphic below.
Run command below to see usage and arguments:
python CTADO.py -husage: Differential analysis of interacting domains between two contact matrices [-h] [-r1 RESULT_DF_1_NAME]
[-r2 RESULT_DF_2_NAME]
[-df RESULT_DATAFRAME_NAME]
[-od OUTPUT_DIRECTORY]
[-nc NUMBER_OF_CHARTS]
[-lg {True,False}] [-t THREADS]
clr1_filename clr2_filename
resolution window flank binsize
clr1_boundaries_name
clr2_boundaries_name
This tool is needed to find four types of changes in TADs between two contact matrices
positional arguments:
clr1_filename Name of first contact matrix in mcool/cool format
clr2_filename Name of second contact matrix in mcool/cool format
resolution Resolution of mcool file
window Size of the sliding diamond window
flank Flank size in bp
binsize Bin size in bp
clr1_boundaries_name The first contact matrix boundaries argument, a dataframe name with TADs boundaries in
chrom, start, end format or cooler insulation table
clr2_boundaries_name The second contact matrix boundaries argument, a dataframe name with TADs boundaries in
chrom, start, end format or cooler insulation table
options:
-h, --help show this help message and exit
-r1 RESULT_DF_1_NAME, --result_df_1_name RESULT_DF_1_NAME
The first contact matrix dataframe name with chrom, start & end of TADs (default: None)
-r2 RESULT_DF_2_NAME, --result_df_2_name RESULT_DF_2_NAME
The second contact matrix dataframe name with chrome, start & end of TADs (default: None)
-df RESULT_DATAFRAME_NAME, --result_dataframe_name RESULT_DATAFRAME_NAME
Dataframe name with intersecting TADs of two contact matrices (default: None)
-od OUTPUT_DIRECTORY, --output_directory OUTPUT_DIRECTORY
The path to the save directory (default: /home/ivandkoz/test_tad_2/differential-computing-
TADs)
-nc NUMBER_OF_CHARTS, --number_of_charts NUMBER_OF_CHARTS
The number of output charts for each type of change. If the specified number is greater than
the number of events, then all of them will be output. If number is -1 than all of them will
be output. (default: 5)
-lg {True,False}, --logging {True,False}
Enables logging (default: False)
-t THREADS, --threads THREADS
Parameter for specifying the number of threads (default: 1)
Good luck! (∿°○°)∿ .・。.・゜✭・.・。.・゜✭・.・。.・゜✭To perform the test run use:
python CTADO.py 4DNFIL6BHWZL_rs.mcool 4DNFIHXCPUAP_rs.mcool 100000 400000 200000 100000 ./data/example/4DNFIL6BHWZL_rs.mcool_400000_boundaries.csv ./data/example/4DNFIHXCPUAP_rs.mcool_400000_boundaries.csv -od ctado_test -nc 3
or
python tads_intensity.py 4DNFIL6BHWZL_rs.mcool 4DNFIHXCPUAP_rs.mcool 100000 400000 200000 100000 4DNFIL6BHWZL_rs.mcool_400000_boundaries.csv 4DNFIHXCPUAP_rs.mcool_400000_boundaries.csv -od ctado_test -r1 4DNFIL6BHWZL_rs.mcool_400000_result_df.csv -r2 4DNFIHXCPUAP_rs.mcool_400000_result_df.csv -df intensity_change_result.csv
Additionaly, you can see the whole example in GitHub Wiki.
Example of graphic:
Intensity
Split
Merge
