TRAPT 0.0.2

A novel deep learning framework for transcription regulators prediction via integraing large-scale epigenomic data.

TRAPT

TRAPT is a novel deep learning framework for transcription regulators prediction via integraing large-scale epigenomic data.

Usage

First, download library:

TRAPT library

Second, install TRAPT:

pip install TRAPT

Run TRAPT using a case:

from TRAPT.TRAPT import Args, RP_Matrix, runTRAPT

# library path
library = 'library'
# input file path
input = 'ESR1@DataSet_01_111_down500.txt'
# output file path
output = 'output/ESR1@DataSet_01_111_down500'

rp_matrix = RP_Matrix(library)
args = Args(input, output)
os.system(f'mkdir -p {output}')
runTRAPT([rp_matrix, args])

Detail

# Constructing TR-RP matrix
python3 CalcTRRPMatrix.py library
# Constructing H3K27ac-RP matrix
python3 CalcSampleRPMatrix.py H3K27ac library
# Constructing ATAC-RP matrix
python3 CalcSampleRPMatrix.py ATAC library
# Reconstruct TR-H3K27ac adjacency matrix
python3 DLVGAE.py H3K27ac library
# Reconstruct TR-ATAC adjacency matrix
python3 DLVGAE.py ATAC library
# Prediction (TR-H3K27ac)-RP matrix
python3 CalcTRSampleRPMatrix.py H3K27ac library
# Prediction (TR-ATAC)-RP matrix
python3 CalcTRSampleRPMatrix.py ATAC library
# TRAPT predicts TR activity
python3 TRAPT.py library input output