Fiona: a parallel and automatic strategy for read error correction.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• High-throughput sequencing data quality is crucial for downstream genomic analyses.
• Existing error correction methods primarily address substitution errors, neglecting indel errors common in technologies like 454 and Ion Torrent.
• Improved read error correction significantly enhances de novo genome assembly and SNP calling.

Purpose of the Study:

• To introduce Fiona, a novel, stand-alone read error correction method.
• To develop a statistical approach for accurate sequencing error detection and correction, with automatic parameter estimation.
• To provide a versatile tool capable of correcting substitution, insertion, and deletion errors for any sequencing technology.

Main Methods:

• Fiona employs a statistical framework for error detection and correction.
• It utilizes an efficient partial suffix array implementation for parallelized read overlap detection with varying seed lengths.
• The method is parameter-free, automatically estimating necessary values from the data.

Main Results:

• Fiona demonstrates superior correction accuracy compared to state-of-the-art methods on diverse real-world datasets.
• The method effectively corrects substitution, insertion, and deletion errors across different sequencing technologies (454, Ion Torrent).
• Fiona achieves high accuracy without compromising computational speed.

Conclusions:

• Fiona is an accurate and parameter-free read error correction tool.
• It is efficient, runnable on inexpensive hardware, and supports multicore parallelization.
• The software is publicly available, facilitating its adoption in genomic research.