Synthetic spike-in standards improve run-specific systematic error analysis for DNA and RNA sequencing

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• High-depth DNA and RNA sequencing can be confounded by systematic sequencing errors (SSEs), which are difficult to distinguish from true biological variants.
• Existing SSE correction methods often rely on recalibration data derived from the target genome or specialized datasets, limiting their applicability.

Purpose of the Study:

• To develop and validate a novel method for improving base quality score recalibration using synthetic spike-in standards.
• To enhance the accuracy of variant calling in complex biological samples by mitigating SSEs.

Main Methods:

• Integration of synthetic RNA spike-in standards into human RNA samples.
• Application of Genome Analysis ToolKit (GATK) for base quality score recalibration using reads mapped to spike-in standards.
• Comparison of recalibration accuracy with conventional GATK recalibration using genome-mapped reads.

Main Results:

• Spike-in based recalibration improved Illumina base quality scores by an average of 5 Phred-scaled units, with up to 13 units improvement at CpG sites.
• Demonstrated overestimation of quality scores for specific dinucleotides (AC, CC, GC, GG, TC) in Illumina RNA sequencing runs.
• Identified distinct SSE patterns for SOLiD sequencing, with fewer dinucleotide SSEs but more cycle-specific errors.

Conclusions:

• Synthetic spike-in standards significantly enhance the accuracy of base quality score recalibration in GATK.
• This method provides run-specific recalibration applicable to diverse species, even those lacking comprehensive SNP databases.
• The findings underscore the utility of spike-in standards for improving the reliability of high-depth sequencing data analysis.