shenscore / htad

Introduction

HTAD is an Active Learning Tool for Matrix-based Detection of Chromatin Domain

Steps to install and run HTAD

(i) Download and install HTAD

$ git clone https://github.com/shenscore/HTAD
$ cd HTAD
$ conda create -n HTAD
$ conda activate HTAD
$ pip install requirements.txt

(ii) calculate the potential TADs and corresponding TAD features

python /path/to/HTAD/src/calcFeatures.py $cooler 10000,20000,40000 outPrefix

then the pickle file of potential TADs will be generated (outPrefix.10k.pkl, outPrefix.20k.pkl, outPrefix.40k.pkl)

(iii) run the HTAD labeler server and train TAD identification model

to start the web server, run:

cd dataLabel
python manage.py runserver

Then visit the corresponding port: e.g. 127.0.0.1:8000

input: the path of your cooler file and feature file of potential TADs, resolution, label(prefix) for current TAD identification model

then the server will select samples (50 for each round) for manual labeling.

The webpage will show corresponding heatmap with TAD marked by yellow triangle.

You can quickly judge whether current TAD is real.

after each round, the server will generate a XX_roundX.h5 model file.

we suggest that the model should be trained in around 10 rounds with the highest resolution to get the best performance.

Note that we could just use the model trained by the highest resolution to predict the real TAD from other resolutions.

(iv) predict the real TADs from potential TADs

given the well trained TAD model file model.h5, run:

python predictTAD.py model.h5 potentialTAD.10k.pkl,potentialTAD.20k.pkl,potentialTAD.40k.pkl 10000,20000,40000 Test

Test.10k.bedpe Test.20k.bedpe Test.40k.bedpe

(v) merge the multi-resolution TAD results

given the sDI value at the highest resolution (e.g. 10kb), run:

python mergeTAD.py 10000 di_check_value.10k 10k.bedpe,20k.bedpe,40k.bedpe final.bedpe

Contact us

Wei Shen: [email protected]