Integrating Hi-C links with assembly graphs for chromosome-scale assembly

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long-read sequencing and long-range assays have transformed de novo genome assembly.
• Hi-C sequencing offers an economical approach for generating chromosome-scale scaffolds.
• Existing open-source Hi-C scaffolding tools have limitations, including higher error rates (inversions, fusions) compared to other methods.

Purpose of the Study:

• To introduce a novel open-source Hi-C scaffolding tool.
• To address limitations in current Hi-C scaffolding methods, specifically error reduction and independence from chromosome number estimation.
• To provide a more accurate and accessible solution for reference genome assembly.

Main Methods:

• Developed a novel Hi-C scaffolder utilizing an assembly graph for error minimization.
• The method does not require an a priori estimate of the chromosome number.
• Implemented in Python and C++ for open-source accessibility.

Main Results:

• Demonstrated higher accuracy compared to state-of-the-art Hi-C scaffolding methods.
• Validated across diverse Hi-C library preparations and input assembly sizes.
• Successfully generated chromosome-scale scaffolds with reduced errors.

Conclusions:

• The novel Hi-C scaffolder provides a more accurate and robust solution for de novo genome assembly.
• The tool's independence from chromosome number estimation and its open-source nature enhance its utility.
• This advancement contributes to the automated reconstruction of high-quality reference genomes.