Accurate circular consensus long-read sequencing improves variant detection and assembly of a human genome

Abstract

The DNA sequencing technologies in use today produce either highly accurate short reads or lessaccurate long reads. We report the optimization of circular consensus sequencing (CCS) to improve the accuracy of single-molecule real-time (SMRT) sequencing (PacBio) and generate highly accurate (99.8%) long high-fidelity (HiFi) reads with an average length of 13.5 kilobases (kb). We applied our approach to sequence the well-characterized human HG002/NA24385 genome and obtained precision and recall rates of at least 99.91% for single-nucleotide variants (SNVs), 95.98% for insertions and deletions <50 bp (indels) and 95.99% for structural variants. Our CCS method matches or exceeds the ability of short-read sequencing to detect small variants and structural variants. We estimate that 2,434 discordances are correctable mistakes in the 'genome in a bottle' (GIAB) benchmark set. Nearly all (99.64%) variants can be phased into haplotypes, further improving variant detection. De novo genome assembly using CCS reads alone produced a contiguous and accurate genome with a contig N50 of >15 megabases (Mb) and concordance of 99.997%, substantially outperforming assembly with less-accurate long reads.

TGS-GapCloser: fast and accurately passing through the Bermuda in large genome using error-prone third-generation long reads

“…There is still a need of error correction or polish for inserted sequences. Note that the most recent Pacbio improved its base-calling accuracy to 99.8% [30], which may directly simplify the problem, but drastically sacrifice the throughout and read length. In this work, we described a software tool, named TGS-GapCloser, that uses errorprone long reads at low coverage to efficiently and accurately close gaps within a reasonable time.…”

Long-read Data Revealed Structural Diversity in Human Centromere Sequences

“…Recently, the advent of long-read sequencing technologies has paved the way for direct, comprehensive observation of sequence variations among various human populations [29][30][31][32][33] . While long-read sequencing was capable of yielding contiguous reference sequences of centromeres for several species 34,35 , reconstruction of whole centromeric sequences for human genomes is still challenging due to their idiosyncratic repeat structures.…”

“…To investigate interindividual variations within the centromeric array, we analyzed single-molecule real-time sequencing reads from four samples: B561 (Japanese), AK1 (Korean) 30 , HG002 (Ashkenazi) 33 , and CHM13 (European) 31 . First, the long reads were preprocessed in silico to filter out the noncentromeric fraction, and the remaining reads were interpreted as a series of alphoid monomers using a catalogue of 1197 monomers, i.e., represented as monomer-encoded reads (Figure 1b).…”

“…SMRTbell libraries were sequenced on Sequel SMRT Cells (Pacific Biosciences) using diffusion loading, 30 kb-insert size and 600-min movies. Other long read data, i.e., from AK1 30 , CHM13 31 , HG002(Hi-Fi) 33 , and CHM13 (ONT ultra-long) 37 were obtained via public repositories respectively.…”

“…In addition to array length, other types of variation are known to exist within an HOR array, such as structurally variant HORs that consist of different numbers and/or types of alpha-satellite monomers 20-23 and also single-nucleotide variations (SNVs) between the HOR units 20, 24, 25 . One of the major driving forces leading to such centromeric variation has been thought to be structural alterations such as unequal crossovers and/or gene conversions 26,27 .Previous studies have investigated centromeric sequence variations via traditional approaches such as restriction enzymes sensitive to alpha-satellite monomers, Southern blotting, or analysis of k-mers unique to centromeres in short reads obtained in the 1000 Genomes Project 28 , but their observations have remained indirect and were confined to specific types of variations due to technological limitations.Recently, the advent of long-read sequencing technologies has paved the way for direct, comprehensive observation of sequence variations among various human populations [29][30][31][32][33] . While long-read sequencing was capable of yielding contiguous reference sequences of centromeres for several species 34,35 , reconstruction of whole centromeric sequences for human genomes is still challenging due to their idiosyncratic repeat structures.…”