Characterization of sequence-specific errors in various next-generation sequencing systems.

Area of Science:

• Genomics
• Bioinformatics
• Microbial Ecology

Background:

• Next-generation sequencing (NGS) is crucial for analyzing microbial diversity, population polymorphisms, and comparative genomics.
• Sequence-specific errors (SSEs) in NGS data can lead to inaccurate genome assembly, inflated diversity metrics, and false polymorphism identification.
• Microbial communities and repetitive genomes are particularly susceptible to SSEs, complicating accurate analysis.

Purpose of the Study:

• To identify and characterize sequence-specific errors (SSEs) across various popular next-generation sequencing (NGS) platforms.
• To investigate the relationship between SSEs and specific sequence features like homopolymers and GC content.
• To develop strategies for mitigating SSEs to improve the accuracy of genomic and microbial community analyses.

Main Methods:

• Utilized a Markov chain model to detect SSEs in public NGS datasets from multiple platforms.
• Analyzed error patterns, focusing on homopolymers and GC content in relation to deletion and substitution errors.
• Applied quality filtering and error correction techniques to mock community data.

Main Results:

• Identified distinct error profiles for different NGS systems; deletion errors often followed homopolymers in non-Illumina systems, while Illumina systems showed substitution errors linked to high GC content and long G/C homopolymers.
• Removing long G/C homopolymers from Illumina HiSeq data improved contig length and SNP quality.
• Quality filtering effectively removed SSEs from mock community data, revealing biases against certain microbes.

Conclusions:

• Characterized SSEs across major NGS platforms, highlighting platform-specific error signatures.
• Demonstrated that targeted filtering, particularly of homopolymers, can significantly enhance data quality and accuracy.
• Provided a foundation for improved bioinformatic pipelines to correct errors, prevent mis-assembly, and ensure accurate microbial community and polymorphism assessments.