Iterative error correction of long sequencing reads maximizes accuracy and improves contig assembly

Area of Science:

• Genomics
• Bioinformatics

Background:

• Next-generation sequencing (NGS) technologies produce long reads (up to 300 bp) beneficial for genome assembly.
• Computational error correction is a crucial first step in genome assembly but remains challenging for longer reads.

Purpose of the Study:

• To develop and evaluate an iterative error correction pipeline for long sequencing reads.
• To improve the accuracy of genome assembly by addressing errors within repetitive regions.

Main Methods:

• An iterative error correction pipeline was developed using modules from the String Graph Assembler (SGA).
• The pipeline employs multiple rounds of k-mer-based correction with increasing k-mer sizes, followed by overlap-based correction.
• This approach combines small and large k-mer advantages to correct errors in repeats.

Main Results:

• The iterative pipeline effectively corrects more errors within repeats compared to standard methods.
• Minimizing erroneous reads leads to a significant increase in contig lengths, two to three times longer.
• Higher read accuracy directly translates to improved genome assembly contiguity.

Conclusions:

• Iterative error correction is a robust strategy for enhancing the quality of long sequencing reads.
• The developed pipeline, SGA-Iteratively Correcting Errors, improves genome assembly by tackling errors in repetitive sequences.
• This method offers a valuable tool for researchers aiming for more contiguous and accurate genome assemblies.