Introduction

TADBpred is a software package designed to evaluate the heterogeneity of topologically associating domain (TAD) boundaries and to predict whether genomic regions contain TAD boundaries. It includes a computational pipeline for integrating TAD boundary data from different TAD samples in a large-scale genomic context. Based on the integrated dataset, TADBpred systematically characterizes TAD boundary heterogeneity, with a focus on variation between active and inactive chromatin. In addition, it provides a machine learning model for TAD boundary prediction by taking boundary heterogeneity into account.

Contents in TADBpred

The TADBpred package contains three main components:

• analysis
  This provides a computational pipeline for integrating TAD boundaries from multiple TAD samples.
• prediction
  This contains the machine learning model used to predict whether a specific genomic region contains a TAD boundary.
• processed_TAD_boundary
  This contains processed TAD boundary datasets generated by the analysis pipeline. It includes two datasets:
  1. A dataset merged from the DomainCaller results
  2. A dataset merged from the InsulationScore results

Software download

Installation

The environment of TADBpred can be installed using Conda:

conda env create -f tad.yml

After installation, activate the environment:

conda activate tad

Citations

If you use the processed TAD boundary dataset or the pipeline for merging TAD boundaries, please cite:

Systematical evaluation of the heterogeneity of topologically associating domain boundaries in large genomic context in human. Sun Y, Tommerup N, Jensen LJ, Gorodkin J*. Submitted.

If you use the tool to predict TAD boundaries, please cite:

Integrating heterogeneity into topologically associating domain boundary prediction in large genomic context in human. Sun Y, Jensen LJ, Tommerup N, Gorodkin J*. Submitted.