Table of contents

• Description
  • Why not package with HyLoRD
• Software Requirements
• Installation
  • Updating
• Usage
  • Job schedulers

Description

SQUIRE (Statistical Quality Utility for Ideal Reference matrix Enhancement) is a python CLI tool that can assist the usage of HyLoRD. The inputs of HyLoRD include a reference matrix and a list of 'useful' CpGs, however the creation of these files is non-trivial.

SQUIRE will:

• Combine bedmethyl files obtained from ONT's modkit into a reference matrix (--reference-matrix).
• Apply statistical tests to each CpG site which can be used for filtering when creating a list of useful CpGs (--cpg-list).
  • Increasing performance considerably

Why not package with HyLoRD

User's might have their own ideas on how to generate the input files for HyLoRD. Maybe they don't have cell sorted ONT data and instead just want to work with WGBS data (more available, but accuracy will suffer). Perhaps they want to work with different statistical tests than the ones used by SQUIRE. For these reasons, SQUIRE has been made independent from HyLoRD as it is completely optional.

Software Requirements

• git
• bash
• python [>=3.12]
• poetry [>=2.1.3]

Installation

First install the project with:

git clone https://github.com/sof202/SQUIRE.git
cd SQUIRE
# Installs to ~/.local/bin
./install.sh 

Alternatively you can install to a user specified location with:

git clone https://github.com/sof202/SQUIRE.git
cd SQUIRE
./install.sh path/to/install/directory

Provided that the installation directory is on your PATH (the installation script will tell you how to do this if this isn't the case), you can run SQUIRE via:

# Print help message for squire
squire --help

Updating

To update SQUIRE:

cd path/to/SQUIRE
git pull
poetry update

Usage

Upon installing SQUIRE successfully, you will have a suite of tools available to you via subcommands of squire. Full details for which can be found via:

squire [subcommand] --help

Typical (bare-bones) usage might be:

squire create -d squire.h5 -b bedmethyl1.bed,bedmethyl2.bed,...,bedmethyln.bed
squire reference -d squire.h5 reference_matrix.bed
squire cpglist -d squire.h5 cpg_list.bed

Further, more in-depth, examples can be found in scripts/.

Job schedulers

It is fairly likely that SQUIRE will take some time to complete statistical tests (especially if the number of samples is high). In the event that the user has access to a HPC (common in bioinformatics) they may want to utilise a job scheduler (such as SLURM, PBS or TORQUE etc.) to run SQUIRE. Example scripts for such uses of SQUIRE can be found in scripts/. These scripts are not comprehensive, but should be a sufficient starting point.