Evaluation of genomic high-throughput sequencing data generated on Illumina HiSeq and Genome Analyzer systems

Abstract

BackgroundThe generation and analysis of high-throughput sequencing data are becoming a major component of many studies in molecular biology and medical research. Illumina's Genome Analyzer (GA) and HiSeq instruments are currently the most widely used sequencing devices. Here, we comprehensively evaluate properties of genomic HiSeq and GAIIx data derived from two plant genomes and one virus, with read lengths of 95 to 150 bases.ResultsWe provide quantifications and evidence for GC bias, error rates, error sequence context, effects of quality filtering, and the reliability of quality values. By combining different filtering criteria we reduced error rates 7-fold at the expense of discarding 12.5% of alignable bases. While overall error rates are low in HiSeq data we observed regions of accumulated wrong base calls. Only 3% of all error positions accounted for 24.7% of all substitution errors. Analyzing the forward and reverse strands separately revealed error rates of up to 18.7%. Insertions and deletions occurred at very low rates on average but increased to up to 2% in homopolymers. A positive correlation between read coverage and GC content was found depending on the GC content range.ConclusionsThe errors and biases we report have implications for the use and the interpretation of Illumina sequencing data. GAIIx and HiSeq data sets show slightly different error profiles. Quality filtering is essential to minimize downstream analysis artifacts. Supporting previous recommendations, the strand-specificity provides a criterion to distinguish sequencing errors from low abundance polymorphisms.

Diagnostic applications of next generation sequencing: working towards quality standards/Diagnostische Anwendung von Next Generation Sequencing: Auf dem Weg zu Qualitätsstandards

“…Overall, these studies show that base substitution errors are more frequent than insertion/deletion type errors, and that the error rate increases in later cycles [14,18,19] . Errors may preferentially occur in specifi c sequence contexts, and therefore errors may be strand specifi c (i.e., occur only in reads mapping to either the forward or the reverse strand of the genome).…”

“…Of note, this study looked at de novo sequencing of a bacterial genome, and it is unclear if the same error rates apply to other applications such as targeted resequencing of human genes. A study evaluating the HiSeq and GAIIx platforms using plant and viral genomes showed comparable error rates, with substitution rates of 0.11 % -0.16 % for the HiSeq and 0.28 % for the older GAIIx instrument, and indel rates of < 3 × 10 -5 [19] . Illumina has stated some basic quality criteria for a successful MiSeq instrument run, these include, for a 2×150 bp run:…”

Snowball: strain aware gene assembly of metagenomes

“…3). As the substitution error rate of the Illumina reads increases towards the ends of the pairedend reads (Minoche et al, 2011), this step results in longer consensus reads with overall lower substitution error, where the overlapping regions are almost error-free. It is also an efficient read quality filtering step, as the paired-end reads that cannot be joined, due to high substitution error rate, an insertion or a deletion within the overlapping region, are filtered out.…”

The genome of an Encephalitozoon cuniculi type III strain reveals insights into the genetic diversity and mode of reproduction of a ubiquitous vertebrate pathogen

“…Read processing, mapping and coverage analysis Quality trimming minimizes downstream artefacts (Minoche et al, 2011) and was performed using the PERL script trim-fastq.pl from the PoPoolation toolkit (Kofler et al, 2011). Quality trimmed reads were used for any downstream analysis.…”