ROCCO: [R]obust [O]pen [C]hromatin Detection via [C]onvex [O]ptimization

What

ROCCO is an efficient algorithm for detection of "consensus peaks" in large datasets with multiple HTS data samples (namely, ATAC-seq), where an enrichment in read counts/densities is observed in a nontrivial subset of samples.

Example Behavior

In the image below, ROCCO is run on a set of ten heterogeneous ATAC-seq samples (lymphoblast) from independent donors (ENCODE). The samples' tracks are colored gray.

• ROCCO consensus peaks (default parameters) are shown in blue
• MACS2 (pooled, q=.01) consensus peaks are shown in red.
• ENCODE cCREs are included as a rough reference of potentially active regions, but note that these regions are not specific to the data samples used in this analysis, nor are they derived from the same cell type or assay.

How

ROCCO models consensus peak calling as a constrained optimization problem with an upper-bound on the total proportion of the genome selected as open/accessible and a fragmentation penalty to promote spatially consistency in active regions and sparsity elsewhere.

Why

ROCCO offers several attractive features:

1. Consideration of enrichment and spatial characteristics of open chromatin signals
2. Scaling to large sample sizes with an asymptotic time complexity independent of sample size
3. No required training data or a heuristically determined set of initial candidate peak regions
4. No rigid thresholds on the minimum number/width of supporting samples/replicates
5. Mathematically tractable model with worst-case bounds on runtime and performance

Paper/Citation

If using ROCCO in your research, please cite the original paper in Bioinformatics (DOI: btad725)

 Nolan H Hamilton, Terrence S Furey, ROCCO: a robust method for detection of open chromatin via convex optimization,
 Bioinformatics, Volume 39, Issue 12, December 2023

Documentation

For additional details, usage, etc. please see ROCCO's documentation: https://nolan-h-hamilton.github.io/ROCCO/

Note that using the module-level functions directly may allow for greater flexibility in applications than using the command-line interface, which is limited in scope.

Installation

PyPI (pip)

pip install rocco # most recent version

pip install --pre rocco # most recent release candidate (if available)

Build from Source

You can also build from source directly if preferred.

• Clone or download this repository

  git clone https://github.com/nolan-h-hamilton/ROCCO.git
  cd ROCCO
  python setup.py sdist bdist_wheel
  pip install -e .

ROCCO utilizes the popular bioinformatics software Samtools and bedtools. If not available already, these system dependencies can be installed with standard MacOS or Linux/Unix package managers, e.g., brew install samtools (Homebrew), sudo apt-get install samtools (APT), Conda conda install bioconda::samtools, etc..

Input/Output

• Input: Samples' BAM alignments or BigWig tracks
• Output: BED file of consensus peak regions