HORmon

HORmon is a tool for annotation of alpha satellite arrays in centromeres of a newly assembled human genome. HORmon consists of two modules:

• Monomer Inference extracts draft human monomers based on the given alpha-satellite consensus template and centromeric sequence.
• HORmon polishes monomers extracted on the previous stage to make it consistent with Centromere Evolution postulate, extract HORs and decompose centromeric sequence into HORs.

HORmon has been used to infer monomers from the recently announced complete human genome assembly of the CHM13 cell line generated by the Telomere-to-Telomere Consortium. The data generated in the paper that describes HORmon (Kunyavskaya et al., 2021) can be found at Figshare. The data includes extracted monomers and HORs from all live alpha satellite arrays in CHM13 cell line, as well as annotations of these arrays. Jupyter notebook that reproduces figures of the HOR paper is available at github.

Installation

The recommended way to install HORmon is with conda package manager:

conda install -c bioconda hormon 

Alternatively, HORmon can be build and installed from source as described below.

Requirements:

• Linux only. Mac OS is not yet supported
• Python3.6+
  • biopython
  • clustalo
  • joblib
  • python-edlib
  • setuptools
  • networkx
  • pygraphviz
  • stringdecomposer

The required python packages can be installed through conda using

conda install --file requirements.txt

Installing from source

git clone https://github.com/ablab/HORmon.git
cd HORmon
python3 setup.py install --record hormon_files.txt

Then, HORmon is available as monomer_inference and HORmon

Afterward, to uninstall HORmon please run

xargs rm -rf < hormon_files.txt

Quick start

Monomer Inference

Monomer Inference script needs two parameters: (1) (centromeric) sequence and (2) monomer template:

monomer_inference -seq test_data/cen8toy.fa -mon test_data/AlphaSat.fa -o toy8_mi

Resulting monomers can be found in toy8_mi/final/monomers.fa and sequence annotation in toy8_mi/final/final_decomposition.tsv.

HORmon

HORmon takes as input: (1) (centromeric) sequence, (2) monomers from "monomer_inference" stage, (3) centromere id for nicer output, (4) the minimum number of occurrence for monomers, monomers pairs and HOR and (5) output folder

HORmon --seq test_data/cen8toy.fa --mon toy8_mi/final/monomers.fa --cen-id 8 --monomer-thr 2 --edge-thr 2 --min-traversals 2 -o toy8

Output:

• toy8/mn.fa -- final monomers
• toy8/final_decomposition.tsv -- monomer decomposition
• toy8/HORs.tsv -- HORs description
• toy8/HORdecomposition.tsv -- HORs decomposition

Note: HORmon should be launch on each centromere independently. We currently cannot guarantee adequate results, in the case of running HORmon on all centromeres simultaneously.

CentromereArchitect

CentromereArchitect (early version of HORmon), as it is described in the paper, is available at the branch centromere-architect. Please cite Dvorkina et al., 2021.

Feedback and bug reports

Your comments, bug reports, and suggestions are very welcomed.
Please leave them at our GitHub repository tracker.

Cite

• Dvorkina, T., Kunyavskaya, O., Bzikadze, A. V., Alexandrov, I., & Pevzner, P. A. (2021). CentromereArchitect: inference and analysis of the architecture of centromeres. Bioinformatics, 37(Supplement_1), i196-i204.
• Kunyavskaya, O., Dvorkina, T, Bzikadze, A. V., Alexandrov, I., & Pevzner, P. A. (2021). HORmon: automated annotation of human centromeres. bioRxiv