Skip to main content
Fig. 2 | Genome Biology

Fig. 2

From: SequencErr: measuring and suppressing sequencer errors in next-generation sequencing data

Fig. 2

Analysis of sequencer errors. a Flow cell-level error rate distribution of common sequencing platforms including HiSeq, NextSeq, and NovaSeq (with the number of flow cells indicated). Outlier flow cells are highlighted in red. Vertical bars indicate the medians. b Flow cell-level error rate across sequencers (instrument identifiers and number of flow cells in parentheses; due to limited availability, sequencers with at least 5, 2, and 1 flow cells are included for HiSeq, NextSeq, and NovaSeq, respectively; see full data in Additional file 1: Fig. S2). Medians are indicated with a vertical black bar and on the right margin of the figure. Red dots indicate outlier sequencers defined at error rate cutoff 100 per million (10−4; vertical dashed line). c Comparison of error rate between the top and bottom surfaces of flow cells. Medians are indicated by vertical bars. Two-sided Wilcoxon rank-sum test P values are indicated for each platform in the right margin. d Prevalence of outlier tiles across flow cells, stratified by the top and bottom surfaces. Shown are percentages (x-axis) of outlier tiles with a high error rate (> 100 per million) for top (blue) and bottom (red) surfaces of each flow cell across three platforms (y-axis). For each platform, the number of flow cells with more than 10% (dashed vertical line) high error rate tiles is indicated by the numbers on the right. *For HiSeq top surfaces, percentages < 1% are replaced with a random number in [0%, 1%] for display purposes. +For HiSeq bottom surfaces, percentage > 30% are replaced with a random number in [30%, 32%] for display purposes

Back to article page