polishCLR: A Nextflow Workflow for Polishing PacBio CLR Genome Assemblies

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long-read sequencing technologies, like Pacific Biosciences (PacBio) continuous long reads (CLR), enable highly contiguous genome assemblies.
• However, CLR data often contains a high error rate, necessitating a polishing step for accurate genome assembly.
• Existing best practices for polishing non-model de novo genome assemblies require a standardized, reproducible workflow for broader accessibility.

Purpose of the Study:

• To develop and present polishCLR, a reproducible Nextflow workflow for polishing genome assemblies generated from CLR data.
• To implement established best practices for error correction in long-read genome assemblies.
• To provide a containerized and publicly available tool for the scientific community.

Main Methods:

• Developed polishCLR, a Nextflow workflow implementing best practices for CLR data polishing.
• Integrated functionalities for handling suboptimal input data and re-entry points for key processes.
• Included steps for duplicate haplotype identification (purge_dups), optional scaffolding, and multiple rounds of polishing and evaluation using Arrow and FreeBayes.

Main Results:

• PolishCLR offers a reproducible workflow for correcting errors in CLR-based genome assemblies.
• The workflow supports various input options and includes re-entry points for flexibility.
• It is containerized and publicly available, facilitating its use in high-performance computing environments.

Conclusions:

• PolishCLR provides a robust and accessible solution for polishing error-prone long-read genome assemblies.
• This workflow enhances the contiguity and accuracy of chromosome-level assemblies.
• It empowers researchers to improve existing genome assemblies derived from CLR data.