Multiplexed Assembly and Annotation of Synthetic Biology Constructs Using Long-Read Nanopore Sequencing

Abstract

Recombinant DNA is a fundamental tool in biotechnology and medicine. These DNA sequences are often built, replicated, and delivered in the form of plasmids. Validation of these plasmid sequences is a critical and time-consuming step, which has been dominated for the last 35 years by Sanger sequencing. As plasmid sequences grow more complex with new DNA synthesis and cloning techniques, we need new approaches that address the corresponding validation challenges at scale. Here we prototype a high-throughput plasmid sequencing approach using DNA transposition and Oxford Nanopore sequencing. Our method, Circuit-seq, creates robust, full-length, and accurate plasmid assemblies without prior knowledge of the underlying sequence. We demonstrate the power of Circuit-seq across a wide range of plasmid sizes and complexities, generating full-length, contiguous plasmid maps. We then leverage our long-read data to characterize epigenetic marks and estimate plasmid contamination levels. Circuit-seq scales to large numbers of samples at a lower per-sample cost than commercial Sanger sequencing, accelerating a key step in synthetic biology, while low equipment costs make it practical for individual laboratories.

Complete sequence verification of plasmid DNA using the Oxford Nanopore Technologies’ MinION device

“…In one particularly long homopolymer from BCRxV.VSVG, only 17.3 % of the reads supported the full 17 bases. Using our data, an existing tool [18] was only able to report 15 of these 17 bases (data not shown).…”

“…This would allow multiple plasmids to be run on the same flowcell, reducing overall cost. Alternatively, barcoding independent plasmids prior to sample loading would allow multiple plasmids to be run at the same time, also reducing costs, as has been done for the related methods described above [14, 18, 19]. While we have benchmarked this procedure on R10.3 flowcells, we fully expect that future developments to this platform (R10.4 and beyond) will only improve the accuracy.…”

“…Existing long read assemblers output a consensus sequence without any per-base quality metrics, and have difficulty processing the large numbers of reads required to gain confidence in the resulting sequence [15][16][17]. Recently, methods have been described that use ONT long read data to validate synthetic plasmid constructs [18,19], focusing on the ability to multiplex plasmids into one sequencing run to reduce cost and increase throughput. While these methods have utility and value, they do not meet the requirements for high-quality plasmid sequence verification, such as in a clinical research setting, where a complete and error-free consensus sequence with interpretable quality scores is needed, and where multiplexing offers a potential source of contamination and uncertainty.…”

Assembly of plant virus agroinfectious clones using biological material or DNA synthesis

“… Alternatives: Alternative facilities and companies commercially providing complete plasmid sequencing include seqWell™ (U.S.A., https://seqwell.com , accessed August 2022), and Plasmidsaurus (U.S.A., https://www.plasmidsaurus.com , accessed August 2022). Note: Sequencing technologies and commercial services suitable for complete plasmid sequencing are rapidly evolving ( Shapland et al., 2015 ; Emiliani et al., 2022 ); it is recommended to periodically evaluate options available in the market. Record the sequences experimentally obtained in in-house and public databases (e.g., NCBI GenBank).…”

“…Note: Sequencing technologies and commercial services suitable for complete plasmid sequencing are rapidly evolving ( Shapland et al., 2015 ; Emiliani et al., 2022 ); it is recommended to periodically evaluate options available in the market.…”