Summarizing and correcting the GC content bias in high-throughput sequencing

Abstract

GC content bias describes the dependence between fragment count (read coverage) and GC content found in Illumina sequencing data. This bias can dominate the signal of interest for analyses that focus on measuring fragment abundance within a genome, such as copy number estimation (DNA-seq). The bias is not consistent between samples; and there is no consensus as to the best methods to remove it in a single sample. We analyze regularities in the GC bias patterns, and find a compact description for this unimodal curve family. It is the GC content of the full DNA fragment, not only the sequenced read, that most influences fragment count. This GC effect is unimodal: both GC-rich fragments and AT-rich fragments are underrepresented in the sequencing results. This empirical evidence strengthens the hypothesis that PCR is the most important cause of the GC bias. We propose a model that produces predictions at the base pair level, allowing strand-specific GC-effect correction regardless of the downstream smoothing or binning. These GC modeling considerations can inform other high-throughput sequencing analyses such as ChIP-seq and RNA-seq.

Resequencing 302 wild and cultivated accessions identifies genes related to domestication and improvement in soybean

“…The aligned read numbers were counted across the whole genome with 200 bp sliding windows and 100 bp slide steps using an in-house Perl script. The GC bias of the Illumina platform was corrected using LOESS smoothing toward a pattern of uniform coverage at all GC percentages, as previously described 54 . CNV candidate windows were initially defined as having five out of seven or more sequential 200 bp overlapping windows with read depth values that differed significantly from the whole-genome average depth (>Mean + 2 × Stdev).…”

No evidence for maintenance of a sympatricHeliconiusspecies barrier by chromosomal inversions

“…The variation is largely due to variation in sequencing depth and PstI site occurrence, which are both related to GC content (Fig. S3; Benjamini and Speed 2012; see Supporting Information note for full discussion). However, SNP density is not correlated with recombination rate, final map lengths, or crossover resolution, and final map lengths are consistent across all crosses (see below), so we do not believe variation in SNP density has affected our results (see Supporting Information note for further details).…”

Assessing Copy Number Alterations in Targeted, Amplicon-Based Next-Generation Sequencing Data

“…The resulting Log 2 (raw copy number ratios) are corrected for the GC content in each amplicon, as previously described (Supplemental Figure S2B), 19 to adjust for the observation that GC-and AT-rich fragments are underrepresented in sequencing due to the unimodal effect that GC content has on DNA melting temperature. This correction is standard for most sequencing-based CNA detection approaches.…”