SequencErr

SequencErr quantifies and suppresses sequencer-induced errors in next-generation sequencing by analyzing base correspondence in overlapping forward and reverse reads to improve accuracy for analyses of heterogeneous cellular populations.

Key Features:

• Error Measurement Methodology: Quantifies sequencer errors by analyzing base correspondence between overlapping regions of forward and reverse reads.
• Comprehensive Error Analysis: Analyzes 3,777 public datasets from 75 research institutions across 18 countries, reporting a typical sequencer error rate of approximately 10 errors per million bases and that 1.4% of sequencers and 2.7% of flow cells exceed 100 errors per million bases.
• Flow Cell Error Profiling: Profiles spatial variability of error rates within flow cells, reporting elevated errors on bottom surfaces and that over 90% of HiSeq and NovaSeq flow cells contain at least one high-error tile.
• Impact on Sequencing Accuracy: Uses sequencing of a common DNA library across different sequencers to show that higher intrinsic sequencer error rates compromise overall accuracy and that excluding outlier error-prone tiles improves data quality.
• Comparative Performance: Compares error metrics to FastQC and to error correction tools Lighter and Musket, reporting roughly a 10-fold reduction in error rate relative to Lighter and Musket.

Scientific Applications:

• Sequencer Performance Assessment: Provides quantitative per-base and per-tile metrics for assessing sequencer and flow-cell performance.
• Error Suppression and Data Quality Improvement: Enables identification and exclusion of outlier high-error tiles to suppress sequencer-induced errors and improve sequencing data quality.
• Benchmarking and QC Comparison: Facilitates benchmarking of sequencing error rates against tools such as FastQC, Lighter, and Musket.
• Analyses of Heterogeneous Cellular Populations: Improves the accuracy of studies targeting heterogeneous cellular populations by reducing instrument-derived errors.
• Calibration and Monitoring for Genomic Research: Supports calibration and continuous monitoring of sequencing accuracy for genomic research and personalized medicine.

Methodology:

Quantifies errors by comparing base calls in overlapping forward and reverse reads; aggregates error rates across 3,777 public datasets; computes per-tile and per-flow-cell spatial error profiles; compares sequencer error metrics across instruments and against FastQC, Lighter, and Musket; and identifies and excludes outlier high-error tiles.