Search

Recombinant plasmid vectors are versatile tools which have facilitated discoveries in molecular biology, genetics, proteomics, and many other fields. As the enzymatic and bacterial processes used to create recombinant DNA can introduce errors, sequence validation is an essential step in plasmid assembly. Sanger sequencing is the current standard for plasmid validation, however this method is limited by an inability to sequence through complex secondary structure and lacks scalability when applied to full-plasmid sequencing of multiple plasmids due to read-length limits. While next-generation sequencing (NGS) does provide full-plasmid sequencing at scale, it is impractical and costly when utilized outside of library-scale validation. Here we present OnRamp (Oxford nanopore-based Rapid Analysis of Multiplexed Plasmids), an alternative method for routine plasmid validation which combines the advantages of NGS’s full plasmid coverage and scalability with Sanger’s affordability and accessibility by leveraging nanopore’s novel long-read sequencing technology. We include customized wet-lab protocols for plasmid preparation along with a pipeline designed for analysis of read data obtained using these protocols. This analysis pipeline is built into the OnRamp webapp (http://OnRamp.BoyleLab.org), which generates alignments between actual and predicted plasmid sequences, quality scores, and read-level views in a user-friendly manner, precluding the need for programming experience in analyzing nanopore results. Here we describe the OnRamp protocols and pipeline, and demonstrate our ability to obtain full sequences from pooled plasmids while detecting sequence variation even in regions of high secondary structure, at less than half the cost of equivalent Sanger sequencing.

Oxford Nanopore sequencers provide results in real time as DNA passes through a nanopore and can eject a molecule after it has been partly sequenced. However, the computational challenge of deciding whether to keep or reject a molecule in real time has limited the application of this capability. We present SquiggleNet, the first deep learning model that can classify nanopore reads directly from their electrical signals. SquiggleNet operates faster than the DNA passes through the pore, allowing real-time classification and read ejection. When given the amount of sequencing data generated in one second, the classifier achieves significantly higher accuracy than base calling followed by sequence alignment. Our approach is also faster and requires an order of magnitude less memory than approaches based on alignment. SquiggleNet distinguished human from bacterial DNA with over 90% accuracy, generalized to unseen species, identified bacterial species in a human respiratory meta genome sample, and accurately classified sequences containing human long interspersed repeat elements.