Resolving repeat families with long reads

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Advancements in long-read sequencing technologies have enabled draft-quality genomes but struggle with complex interspersed repeat families.
• Current genome assemblies are often fragmented at contig and scaffold ends due to unresolved repetitive elements.
• Existing algorithmic solutions for repeat resolution are limited by scalability and high read error rates inherent in long-read data.

Purpose of the Study:

• To develop and assess novel computational methods for resolving large interspersed repeat families in genome assemblies.
• To improve the contiguity and completeness of genome assemblies, particularly for complex genomes.

Main Methods:

• Proposed novel repeat resolution algorithms designed to handle large copy numbers and high read error rates.
• Evaluated method accuracy using simulated datasets with diverse repeat structures.
• Tested performance on *Drosophila melanogaster* transposons and compared results against an existing long-read repeat resolution tool.

Main Results:

• The proposed methods demonstrate improved accuracy in resolving interspersed repeat families compared to existing tools.
• Successful application on simulated data and *Drosophila melanogaster* transposons highlights the robustness of the approach.
• Validation shows enhanced ability to handle challenges posed by high error rates and multiple repeat copies.

Conclusions:

• The developed methods are effective for improving the contiguity of genome assemblies.
• This work provides a significant advancement for assembling complex genomes using long-read sequencing data.
• The findings pave the way for more complete and accurate genomic blueprints across diverse organisms.