Email updates

Keep up to date with the latest news and content from BMC Bioinformatics and BioMed Central.

Open Access Research article

Error-correcting properties of the SOLiD Exact Call Chemistry

Tim Massingham and Nick Goldman

Author Affiliations

European Bioinformatics Institute, Wellcome Trust Genome Campus, Hinxton, Cambridgeshire, UK

BMC Bioinformatics 2012, 13:145  doi:10.1186/1471-2105-13-145

Published: 22 June 2012

Abstract

Background

The Exact Call Chemistry for the SOLiD Next-Generation Sequencing platform augments the two-base-encoding chemistry with an additional round of ligation, using an alternative set of probes, that allows some mistakes made when reading the first set of probes to be corrected. Additionally, the Exact Call Chemistry allows reads produced by the platform to be decoded directly into nucleotide sequence rather than its two-base ‘color’ encoding.

Results

We apply the theory of linear codes to analyse the new chemistry, showing the types of sequencing mistakes it can correct and identifying those where the presence of an error can only be detected. For isolated mistakes that cannot be unambiguously corrected, we show that the type of substitution can be determined, and its location can be narrowed down to two or three positions, leading to a significant reduction in the the number of plausible alternative reads.

Conclusions

The Exact Call Chemistry increases the accuracy of the SOLiD platform, enabling many potential miscalls to be prevented. However, single miscalls in the color sequence can produce complex but localised patterns of error in the decoded nucleotide sequence. Analysis of similar codes shows that some exist that, if implemented in alternative chemistries, should have superior performance.