Recent progress in atomistic simulation of electrical current DNA sequencing

Highlights

• Recent advances in the transverse electrical current DNA sequencing are reviewed.

• The DNA sequencers are categorized according to their sensing mechanisms.

• Devices based on low dimensional nanomaterials are emphasized .

• Particular emphasis is placed on modeling and simulation.

• Issues in computational studies and desirable research directions are summarized.

Abstract

We review recent advances in the DNA sequencing method based on measurements of transverse electrical currents. Device configurations proposed in the literature are classified according to whether the molecular fingerprints appear as the major (Mode I) or perturbing (Mode II) current signals. Scanning tunneling microscope and tunneling electrode gap configurations belong to the former category, while the nanochannels with or without an embedded nanopore belong to the latter. The molecular sensing mechanisms of Modes I and II roughly correspond to the electron tunneling and electrochemical gating, respectively. Special emphasis will be given on the computer simulation studies, which have been playing a critical role in the initiation and development of the field. We also highlight low-dimensional nanomaterials such as carbon nanotubes, graphene, and graphene nanoribbons that allow the novel Mode II approach. Finally, several issues in previous computational studies are discussed, which points to future research directions toward more reliable simulation of electrical current DNA sequencing devices.

Introduction

The sequencing of DNA not only has significant scientific implications in the context of deciphering the fundamental code of life but also represents an enormous opportunity to improve the well-being of humankind by ushering in a new era of personal or precision medicine (Mardis, 2011, Shendure and Aiden, 2012, Rabbani et al., 2014). A decade after the completion of the Human Genome Project in 2003, effort is now devoted into the development of next-generation DNA sequencing technologies that can meet the ‘$1000 genome’ goal set by National Institute of Health. In this endeavor, in contrast to the second-generation DNA sequencing technologies that still require polymerase chain reaction amplification and fluorescent labeling as in the first-generation counterpart, the newly-emerged third-generation DNA sequencing technologies propose single molecule detection based on changes in ionic or electrical currents.

The fundamental ingredient of the third-generation DNA sequencing technology is a nanopore, through which a double-stranded or single-stranded DNA translocates in a linear conformation when it is driven by an electric field (Branton et al., 2008, Guo et al., 2014, Wang et al., 2015, Yokota et al., 2014, Zwolak and Di Ventra, 2008). Nanopore-based DNA sequencing was first proposed for protein pores such as α-hemolysin (Kasianowicz et al., 1996) and MspA (Butler et al., 2008). In the biological nanopore approach, trans-membrane longitudinal-direction ionic current blockades are monitored as schematically shown in Fig. 1a. While the biological nanopore approach has advantages such as the well-defined atomic-scale pore size and corresponding good signal-to-noise ratio, it also suffers from problems such as the instability of lipid membranes in an electric field.

As an alternative to biological nanopores, solid-state nanopores have been actively investigated for the next-generation DNA sequencing. Within this scheme, in addition to the longitudinal-direction ionic current, another fundamentally different sequencing mode, i.e. the transverse-direction electrical current is available (Fig. 1b). The solid-state nanopore DNA sequencing approach based on transverse electrical currents will allow employing advanced semiconductor device fabrication techniques that are well-established in the microelectronics industry. It has also attracted much attention in that novel low-dimensional nanomaterials such as carbon nanotubes (CNTs), graphene, and graphene nanoribbons (GNRs) can be adopted as electrodes and/or nanopores, which will potentially provide novel device geometries and improved device characteristics.

The two core technologies of the transverse electrical current DNA sequencing approach are (1) reading and distinguishing nucleobases at the single-molecule level (sensing and electronics) and (2) understanding and controlling the DNA translocation dynamics (nanofluidics). Theory and computation have been playing an important role in both area by proposing various novel reading mechanisms and providing atomistic pictures of translocation processes. Indeed, the concept itself was theoretically proposed first along the line of molecular electronics that involve metal electrodes (Lee and Thundat, 2005, Zwolak and Di Ventra, 2005). Since then, proof-of-principles experiments using Au electrodes have been successfully carried out (Chang et al., 2010, Huang et al., 2010, Ohshiro et al., 2012, Tsutsui et al., 2010). The employment of novel non-metal low-dimensional materials such as CNTs (Chen et al., 2012, Kim et al., 2014, Meng et al., 2006, Meunier and Krstić, 2008, Sadeghi et al., 2014a), graphene (Ahmed et al., 2012, Postma, 2010, Prasongkit et al., 2011, Prasongkit et al., 2013), GNRs (Ahmed et al., 2014b, Avdoshenko et al., 2013, Cho et al., 2011, Girdhar et al., 2013, Girdhar et al., 2014, He et al., 2011, Jeong et al., 2013, Min et al., 2011, Nelson et al., 2010, Rajan et al., 2014, Rezapour et al., 2014, Saha et al., 2012, Shenglin et al., 2014, Zhang et al., 2014, Zhao et al., 2012), and other low-dimensional materials were also first proposed by theoreticians (Amorim and Scheicher, 2014, Farimani et al., 2014, He et al., 2014, Sadeghi et al., 2014a, Sadeghi et al., 2014b, Thomas et al., 2014), and along this line significant experimental advances are currently being made (Chen et al., 2013, Garaj et al., 2010, Merchant et al., 2010, Schneider et al., 2010, Traversi et al., 2013).

In this article, we critically review recent progress made in the computational study of DNA sequencing based on the detection of transverse electrical currents. Referring other review papers that have extensively discussed the DNA translocation dynamics and control (Fyta et al., 2011, Luan et al., 2012, Zwolak and Di Ventra, 2008), we will particularly focus on the single DNA sensing mechanisms together with the utilization of novel low-dimensional materials and corresponding device geometries. In Table 1, we summarized the representative theoretical literature in terms of electrode materials, device geometries, modeling features, simulation levels, signal type and ranking, and reading mechanisms.

The organization of this article is as follows: In Section 2, we first summarize the theoretical formulation for the calculation of electrical currents in nanoscale junctions. Proposed device configurations and sensing mechanisms are categorized according to whether molecular signals arise from the tunneling current (Mode I) or electrochemical gating effect (Mode II). In the remainder, we will divide the discussion in terms of inorganic materials used as the probe electrode in the DNA sequencer. Basic features in the electronic and transport properties of CNTs, graphene, and GNRs will be also summarized. In Section 3, we first discuss the DNA sequencing approaches employing metal nanoelectrodes that can be used as a scanning tunneling microscope (STM) tip or nanogap electrodes (Mode I), for which proof-of-principles experiments have been successfully carried out. A big advantage of employing recently emerged low-dimensional nanomaterials such as CNTs and graphene is that it can potentially allow a novel DNA sequencing mechanism (Mode II). In 4 DNA sequencing based on carbon nanotubes, 5 DNA sequencing based on graphene and graphene nanoribbons, we will review computational studies that have considered CNTs and graphene, respectively. In addition to CNTs and graphene, a whole new family of novel two-dimensional nanomaterials such as hexa-boron nitride (hBN), silicene, and transition metal dichalcogenides (e.g. MoS₂) have been attracting great attention, and now several studies have appeared on their applications to the DNA sequencing (Section 6). We will also address potential issues in previous computational studies (Section 7), which naturally lead us to future research directions (Section 8).