This is a standalone repository for the local command line version of BART (v2.1.1).

For the source code of BART web interface, please refer to:

• BARTweb Frontend for receiving users’ job submission requests and displaying job execution information and results.
• BARTweb Backend for performing all of the computation.

For the custom code for processing the data and generating the figures in the manuscript of BARTweb: a web server for transcriptional regulator association analysis, please refer to:

• BARTweb custom code

BART (Binding Analysis for Regulation of Transcription) is a bioinformatics tool for predicting functional transcriptional regulators (TRs) that bind at genomic cis-regulatory regions to regulate gene expression in the human or mouse genomes, taking a query gene set, a ChIP-seq dataset or a scored genomic region set as input. BART leverages over 7,000 human TR binding profiles and over 5,000 mouse TR binding profiles from the public domain (collected in Cistrome Data Browser) to make the prediction.

BART is implemented in Python and distributed as an open-source package along with necessary data libraries.

BART web interface (Beta version) can be accessed here.

BART is developed and maintained by the Chongzhi Zang Lab at the University of Virginia.

BART requires Python 3 and the following packages. We highly recommend using a conda environment or virtual environment.

• numpy
• pandas
• scipy
• tables
• scikit-learn
• matplotlib

You have to download the Human or Mouse Data Library before using BART. The unpacked libraries occupy 14GB storage.

wget https://virginia.box.com/shared/static/2kqczz9gixetcr9p4bl650uyrio5zd33.gz -O hg38_library.tar.gz
tar zxf hg38_library.tar.gz
wget https://virginia.box.com/shared/static/bxdggnhp4bjz2l5h2zjlisnzp0ac7axf.gz -O mm10_library.tar.gz
tar zxf mm10_library.tar.gz

hg38:https://zenodo.org/records/18854649/files/hg38_library.tar.gz?download=1

mm10:https://zenodo.org/records/18854649/files/mm10_library.tar.gz?download=1

hg38:https://myuva-my.sharepoint.com/:u:/g/personal/hz9fq_virginia_edu/IQB2IqcSn23wSaVIP9PoUS1iAVgA5x4T06AzsBcrQ0wLiDA?e=RPOgde

mm10:https://myuva-my.sharepoint.com/:u:/g/personal/hz9fq_virginia_edu/IQCankHDq3WqQIO7zYFP3EiqAe_bjFK14kWsQ8kJIpYOJZg?e=pO5PK1

Clone the latest version of BART from github.

git clone https://github.com/zanglab/bart2.git
cd bart2

Or, download a source distribution of BART and go to the directory where you unpacked BART.

wget https://virginia.box.com/shared/static/jvwc097d7ca0oart6ka4ivonqi4qxkox.zip -O bart2.zip
unzip bart2.zip
cd bart2

Backup URLs for source code: https://zenodo.org/records/19341963/files/bart2.zip?download=1

Then you have to modify the configure file (bart2/bart.conf). For example, if you have the hg38_library (or mm10_library) downloaded in this directory: /abc/def/hg38_library (or /abc/def/mm10_library), then the bart.conf file should read:

[path]
hg38_library_dir = /abc/def/
mm10_library_dir = /abc/def/

Install with root/administrator permission, or you have the Anaconda environment prepared. By default, the script will install python library and executable codes globally.

pip install .

If you want to install everything under a specific directory, for example, a directory as /path/to/bart2/, use the following commands.

mkdir -p /path/to/bart/lib/pythonX.Y/site-packages 
export PYTHONPATH=/path/to/bart/lib/pythonX.Y/site-packages/:$PYTHONPATH 
pip install . -t /path/to/bart
export PATH=/path/to/bart/bin/:$PATH

In this value, X.Y stands for the major–minor version of Python you are using (such as 3.5 ; you can find this with sys.version[:3] from a Python command line).

You’ll need to modify the environment variables and add those lines in your bash file (varies on each platform, usually is ~/.bashrc or ~/.bash_profile).

export PYTHONPATH= "/path/to/bart/lib/pythonX.Y/site-packages/:$PYTHONPATH"
export PATH="/path/to/bart/bin/:$PATH"

{geneset, profile, region}

Given a query gene set in official gene symbols (HGNC for human or MGI for mouse) in text format (each gene in a row, at least 100 genes recommended), predict functional TRs that regulate these genes.

Usage: bart2 geneset -i genelist.txt -s hg38 --outdir bart2_output

Given a ChIP-seq data file (mapped reads in BAM or BED format in either hg38 or mm10), predict TRs whose binding pattern associates with the input ChIP-seq profile.

Usage: bart2 profile -i ChIP.bam -f bam -s hg38 --outdir bart2_output

Given a scored genomic region set (BED format in either hg38 or mm10), predict TRs enriched in this genomic region set.

Usage: bart2 region -i ChIPpeak.bed -c 4 -s hg38 --outdir bart2_output

1. *_adaptive_lass_Info.txt provides regression information tells which representative H3K27ac samples are selected along with coefficients through adaptive lasso regression and sample annotations including cell line, cell type or tissue type. This is the output only generated in geneset mode.

2. *_CRE_prediction_lasso.txt is the predicted cis-regulatory profile of the input gene set and is a ranked list of all CREs (UDHS) in the genome. The higher the score, the more likely the regulatory element regulates the input gene set. This is the output only generated in geneset mode.

3. *_auc.txt provides the association score of each of the TR ChIP-seq dataset with the genome cis-regulatory profile.

4. *_bart_results.txt is a rank of all TRs with multiple quantification scores.

Please refer here for an example of BART2 results.

If you use BART in your data analysis, please cite:

BART: a transcription factor prediction tool with query gene sets or epigenomic profiles
Zhenjia Wang, Mete Civelek, Clint Miller, Nathan Sheffield, Michael J. Guertin, Chongzhi Zang. Bioinformatics 34, 2867–2869 (2018)