Characterizing and measuring bias in sequence data

By M. Ross, C. Russ, Maura Costello, A. Hollinger, Niall J. Lennon, Ryan Hegarty, Chad Nusbaum, David B. Jaffe. Genome Biology, 2013

913 citations recorded by the source corpus.

DNA sequencing technologies deviate from the ideal uniform distribution of reads. These biases impair scientific and medical applications. Accordingly, we have developed computational methods for discovering, describing and measuring bias. We applied these methods to the Illumina, Ion Torrent, Pacific Biosciences and Complete Genomics sequencing platforms, using data from human and from a set of microbes with diverse base compositions. As in previous work, library construction conditions significantly influence sequencing bias. Pacific Biosciences coverage levels are the least biased, followed by Illumina, although all technologies exhibit error-rate biases in high- and low-GC regions and at long homopolymer runs. The GC-rich regions prone to low coverage include a number of human promoters, so we therefore catalog 1,000 that were exceptionally resistant to sequencing. Our results indicate that combining data from two technologies can reduce coverage bias if the biases in the component technologies are complementary and of similar magnitude. Analysis of Illumina data representing 120-fold coverage of a well-studied human sample reveals that 0.20% of the autosomal genome was covered at less than 10% of the genome-wide average. Excluding locations that were similar to known bias motifs or likely due to sample-reference variations left only 0.045% of the autosomal genome with unexplained poor coverage. The assays presented in this paper provide a comprehensive view of sequencing bias, which can be used to drive laboratory improvements and to monitor production processes. Development guided by these assays should result in improved genome assemblies and better coverage of biologically important loci.

Subfields: Genetics, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/8c77c718563b5e823feac3a4e0b843c75457f5fd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Genetics

In Molecular Biology

Citations in the corpus

SVIM: structural variant identification using mapped long reads (2018, background): “While PacBio produces more insertion than deletion errors [37], Nanopore shows the opposite tendency [38].”
NOVOPlasty: de novo assembly of organelle genomes from whole genome data (2016, methodology): “Current Illumina sequencing technology (Illumina HiSeq and MiSeq) has for example a very high error rate after a long single nucleotide repeat (SNR) stretch, which makes the subsequent sequence unreliable (16).”
Mapping and elucidating the function of modified bases in DNA (2017, methodology): “Tomkova, M., McClellan, M., Kriaucionis, S. & Schuster-Boeckler, B. 5- hydroxymethylcytosine marks regions with reduced mutation frequency in human DNA.”
Assembling large genomes with single-molecule sequencing and locality-sensitive hashing (2014, methodology): “The raw reads are then corrected using a consensus of overlapping reads, and the corrected reads are assembled using an updated version of Celera Assembler 13 .”
Identifying conserved genomic elements and designing universal bait sets to enrich them (2017, background): “…(~400–500 MYA; Sanders and Lee 2010) relative to the other arthropod groups (~150–250 MYA; Misof et al. 2014), and (ii) account for sequencing error (Ross et al. 2013) in the arachnid Illumina data relative to the simulated reads generated for other arthropod taxa (which derive from a putatively…”
Canu: scalable and accurate long-read assembly via adaptive k-mer weighting and repeat separation (2016, background): “Nevertheless,PacBio single-molecule real-time (SMRT)sequencing exhibits a largely unbiased and random error model (Ross et al. 2013), enabling assemblies that exceed short-read data both in terms of quality and continuity (Chin et al. 2013; Koren et al. 2013).”
Computational pan-genomics: status, promises and challenges (2016, background): “Third, there is a nonuniform distribution of sequencing coverage [81] owing to various factors including biases in polymerase chain reaction amplification, polymerase processivity and bridge amplification.”
Rare-variant association analysis: study designs and statistical tests. (2014, background): “90 Careful variant ﬁltering and quality control are important for avoiding the identiﬁcation of associations that are driven by between-platform heterogeneity.”
NOVOPlasty: de novo assembly of organelle genomes from whole genome data. (2016, methodology): “Current Illumina sequencing technology (Illumina HiSeq and MiSeq) has for example a very high error rate after a long single nucleotide repeat (SNR) stretch, which makes the subsequent sequence unreliable (16).”
proovread: large-scale high-accuracy PacBio correction through iterative short read consensus (2014, background): “%) in a random pattern (Ross et al., 2013).”
RNA Sequencing Data: Hitchhiker's Guide to Expression Analysis (2018, background): “By sequencing single molecules, they can also avoid the need for PCR amplification, hence reducing coverage biases [207,208].”
Sequencing DNA with nanopores: Troubles and biases (2021, background): “The relative coverage [29] is normally defined as the coverage of a given reference base in a genome divided by the mean coverage of all reference bases.”