Duplex Proximity Sequencing (Pro-Seq): A method to improve DNA sequencing accuracy without the cost of molecular barcoding redundancy

Abstract

13A challenge in the clinical adoption of cell-free DNA (cfDNA) liquid biopsies for cancer 14 care is their high cost compared to potential reimbursement. The most common approach 15 used in liquid biopsies to achieve high specificity detection of circulating tumor DNA 16 (ctDNA) among a large background of normal cfDNA is to attach molecular barcodes to 17 each DNA template, amplify it, and then sequence it many times to reach a low-error 18 consensus. In applications where the highest possible specificity is required, error rate can 19 be lowered further by independently detecting the sequences of both strands of the starting 20 cfDNA. While effective in error reduction, the additional sequencing redundancy required 21 by such barcoding methods can increase the cost of sequencing up to 100-fold over 22 standard next-generation sequencing (NGS) of equivalent depth. 23We present a novel library construction and analysis method for NGS that achieves 24 comparable performance to the best barcoding methods, but without the increase in 25 sequencing and subsequent sequencing cost. Named Proximity-Sequencing (Pro-Seq), the 26 method merges multiple copies of each template into a single sequencing read by 27 physically linking the molecular copies so they seed a single sequencing cluster. Since 28 multiple DNA copies of the same template are compared for consensus within the same 29 cluster, sequencing accuracy is improved without the use of redundant reads. Additionally, 30 it is possible to represent both senses of the starting duplex in a single cluster. The resulting 31 workflow is simple, and can be completed by a single technician in a work day with 32 minimal hands on time. 33Using both cfDNA and cell line DNA, we report the average per-mutation detection 34 threshold and per-base analytical specificity to be 0.003% and >99.9997% respectively, 35 demonstrating that Pro-Seq is among the highest performing liquid biopsy technologies in 36 terms of both sensitivity and specificity, but with greatly reduced sequencing costs 37 compared to existing methods of comparable accuracy. 38

Enhancing the accuracy of next-generation sequencing for detecting rare and subclonal mutations

“…Together these features can improve consensus-making efficiency by an order of magnitude. A clever approach to eliminate amplicon competition and normalize yield per founding molecule using digital emulsion amplification has been demonstrated with both PCR and rolling-circle amplification 71,77,102 . Ultimately, the most efficient consensus sequencing approaches are those that do not require any amplification, but single-molecule technologies will need to improve considerably before this becomes a major consideration.…”

“…yellow triangles in Figure 2e). A recent preliminary technique known as Pro-Seq involves concurrent amplification of both the reference and anti-reference strands of individual DNA duplexes with physically linked primers in emulsion droplets 77 . This enables genotyping of both strands within the same flow-cell cluster for cost savings, but has an identical limitation in that amplification of both strands cannot be ensured.…”

“…In theory a variety of other molecular tools could serve as Duplex Sequencing UMIs and SDEs. Other than shear points and DNA-based tags, single-molecule compartmentalization methods that keep paired strands in physical proximity 77 or other non-nucleic acid tagging methods could serve the strand-relating function. Similarly, assymetric chemical labelling of the adapter strands in a way that they can be physically separated can serve an SDE role.…”

Detecting Rare Mutations and DNA Damage with Sequencing-Based Methods

“…For example, even when optimized, some of these sequencing methods result in a ~40-fold reduction of usable consensus sequence relative to the raw quantity of sequence data that are generated [19]. As described above, one means to address this shortcoming is to physically link (i.e., concatenate) amplified copies of the same original DNA fragment in a single library molecule, using methods such as circle sequencing [1, 33, 34], o2n-Seq [22], or Pro-Seq [37] (Figure 1b). By doing so, each individual sequence read provides redundancy without the inefficiency of randomly sampling from a pool of overamplified library molecules, resulting in higher yields.…”

“…One disadvantage of circle sequencing is that the RCA step creates another opportunity for biased amplification, exacerbating variation in coverage across a genomic sample. Therefore, obtaining more even representation through methods such as emulsion PCR has been a goal of recent modifications to sequencing protocols [22, 34, 37]. Overall, circle sequencing and related methods that generate physically linked read families are the preferred means to obtain the largest possible yields of consensus sequence.…”