D3 is a fast and accurate tool for computing DNA Density and Distance to periphrery (DisTP). In our paper, we show that the DNA density and DisTP are highly correlated with nuclear activities.
paper.
D3 was tested on Python v3.9.1 (macOS and DeepinOS), with the following basic requirements:
- NumPy (tested on v1.20.2)
- SciPy (tested on v1.6.2)
- Pandas (tested on v1.2.4)
- Seaborn (tested on v0.11.1)
- Matplotlib (tested on v3.3.4)
Below is a typical workflow using the test data.
D3.py D3 and D3.py D3s compute the DNA density and DisTP. The resulted DNA and DisTP are stored in bed-like format file.
cd PATH/WHERE/D3/AT
mkdir -p test_result/den_dtp
python D3.py D3s test_data/dg_files test_data/hg19_diplo_20k.window.bed test_result/den_dtp
D3.py sta gives the density and DisTP ranges, and a scatter plot as below.
D3.py map puts the bins on density-DisTP matrix, and stores the probability of genomic bins appearing at matrix bins (states).
D3.py ave computes the average and standard deviation (sd) of density and DisTP.
python D3.py sta test_result/den_dtp/den_dtp test_data/hg19_diplo_20k.window.bed test_result/test_map_sta
python D3.py map test_result/den_dtp/den_dtp test_data/hg19_diplo_20k.window.bed test_result/test_map
python D3.py ave test_map.txt test_ave.txt
D3.py marks indexes and concatenates the markers. D3.py enrich plotted the enrichments of markers individually. D3.py hiera ranks the matrix bins (states) by hierarchy cluster.
python D3.py marks test_data/marks/ test_data/hg19_diplo_20k.window.bed test_mark.mark.txt
mkdir mark_enrich
python D3.py enrich test_map.txt test_mark.mark.txt mark_enrich/gm12878
python D3.py hiera test_map.txt test_mark.mark.txt test_hiera
Markers Enrichments: gm12878_1-Active-Promoter_histplot.pdf gm12878_13-Heterochrom_histplot.pdf
Hierarchy cluster: test_hiera_value_hierarchy.pdf test_hiera_hierarchy_hist.pdf
D3 is free for non-commercial use by academic, government, and non-profit/not-for-profit institutions. A commercial version of the software is available and licensed through Xi’an Jiaotong University. For more information, please contact with Yizhuo che ([email protected]) or Kai Ye ([email protected]).