Skip to content

JiekaiLab/iceDP

Repository files navigation

iceDP project description

The iceDP package processes genome interaction data using a density peaks algorithm to identify inter-chromatin interactions. Below is the procedure for using iceDP effectively.


Installation

Use pip to install:

pip install iceDP

Or directly copy source code:

# copy to:
${python_path}/site-packages

Usage

  • Data required

iceDP receive any genome interaction data in a three-column format as input. Before running analyses, we need to prepare genome interaction data in a three-column format:

Column 1 (chr_one_site) → Genomic position on the first chromosome.

Column 2 (chr_two_site) → Genomic position on the second chromosome.

Column 3 (interaction_value) → Interaction strength between the two positions (typically observed in Hi-C experiments).

Check the sample data:

head -5 play_data/chr4_chr11_mm10.txt
3090000 3100000 1.0
3795000 3100000 1.0
4205000 3100000 1.0
4255000 3100000 1.0
4230000 3105000 1.0

This file represents genomic interactions between chromosome 4 and chromosome 11 in the mouse genome (mm10).

If your data is stored in .hic format, you can extract interaction values using Juicer Tools:

java -jar juicer_tools.jar dump observed NONE yourdata.hic chr4 chr11 BP 50000 chr4_chr11_mm10.txt

  • iceDP procedure
import iceDP

# Initialize data processing
x = iceDP.main_procedure.bunchDots()

# Read interaction data from file
x.readData('play_data/chr4_chr11_mm10.txt')

# Compute parameter for decision tree
x.get_rho()    # Calculate rho values
x.get_delta()  # Calculate delta values

# Statistical analysis of interaction sites
x.do_chi_square_test()  # Perform chi-square test
x.define_border()       # Define interaction borders
x.horizontal_and_vertical_fold_change()  # Fold change for horizontal and vertical

# Save the results to a formatted output file
iceDP.main_procedure.save_reult(x, 'chr4_chr11_mm10_DPresult')

  • Output format

After processing, iceDP generates a comma-separated 15-column table, filtering out low-confidence interaction spots. Each row in the saved output file (chr4_chr11_mm10_DPresult) contains the following columns:

Column Name Description
index Spot identifier
start1 Position on chromosome 1
start2 Position on chromosome 2
counts Interaction counts
rho Density metric
delta Peak distance metric
p_value Statistical significance (Chi-square test)
fillN Number of square near center with counts
center_x Center of interaction site (X)
center_y Center of interaction site (Y)
side_length_x Interaction region size (X axis)
side_length_y Interaction region size (Y axis)
fold_change Center square to surround square fold change
vertical_fc Vertical axis fold change
horizontal_fc Horizontal axis fold change

  • Plot
iceDP.plot_spots.plot_one_spot(x.data_filted2.values[1], x)

plot result

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages