A machine learning approach for accurate and real-time DNA sequence identification.

BACKGROUND: The all-electronic Single Molecule Break Junction (SMBJ) method is an emerging alternative to traditional polymerase chain reaction (PCR) techniques for genetic sequencing and identification. Existing work indicates that the current spectra recorded from SMBJ experimentations contain unique signatures to identify known sequences from a dataset. However, the spectra are typically extremely noisy due to the stochastic and complex interactions between the substrate, sample, environment, and the measuring system, necessitating hundreds or thousands of experimentations to obtain reliable and accurate results. RESULTS: This article presents a DNA sequence identification system based on the current spectra of ten short strand sequences, including a pair that differs by a single mismatch. By employing a gradient boosted tree classifier model trained on conductance histograms, we demonstrate that extremely high accuracy, ranging from approximately 96 % for molecules differing by a single mismatch to 99.5 % otherwise, is possible. Further, such accuracy metrics are achievable in near real-time with just twenty or thirty SMBJ measurements instead of hundreds or thousands. We also demonstrate that a tandem classifier architecture, where the first stage is a multiclass classifier and the second stage is a binary classifier, can be employed to boost the single mismatched pair's identification accuracy to 99.5 %. CONCLUSIONS: A monolithic classifier, or more generally, a multistage classifier with model specific parameters that depend on experimental current spectra can be used to successfully identify DNA strands.

Identifying SARS-CoV-2 Variants Using Single-Molecule Conductance Measurements.

The global COVID-19 pandemic has highlighted the need for rapid, reliable, and efficient detection of biological agents and the necessity of tracking changes in genetic material as new SARS-CoV-2 variants emerge. Here, we demonstrate that RNA-based, single-molecule conductance experiments can be used to identify specific variants of SARS-CoV-2. To this end, we (i) select target sequences of interest for specific variants, (ii) utilize single-molecule break junction measurements to obtain conductance histograms for each sequence and its potential mutations, and (iii) employ the XGBoost machine learning classifier to rapidly identify the presence of target molecules in solution with a limited number of conductance traces. This approach allows high-specificity and high-sensitivity detection of RNA target sequences less than 20 base pairs in length by utilizing a complementary DNA probe capable of binding to the specific target. We use this approach to directly detect SARS-CoV-2 variants of concerns B.1.1.7 (Alpha), B.1.351 (Beta), B.1.617.2 (Delta), and B.1.1.529 (Omicron) and further demonstrate that the specific sequence conductance is sensitive to nucleotide mismatches, thus broadening the identification capabilities of the system. Thus, our experimental methodology detects specific SARS-CoV-2 variants, as well as recognizes the emergence of new variants as they arise.

Mapping DNA Conformations Using Single-Molecule Conductance Measurements.

DNA is an attractive material for a range of applications in nanoscience and nanotechnology, and it has recently been demonstrated that the electronic properties of DNA are uniquely sensitive to its sequence and structure, opening new opportunities for the development of electronic DNA biosensors. In this report, we examine the origin of multiple conductance peaks that can occur during single-molecule break-junction (SMBJ)-based conductance measurements on DNA. We demonstrate that these peaks originate from the presence of multiple DNA conformations within the solutions, in particular, double-stranded B-form DNA (dsDNA) and G-quadruplex structures. Using a combination of circular dichroism (CD) spectroscopy, computational approaches, sequence and environmental controls, and single-molecule conductance measurements, we disentangle the conductance information and demonstrate that specific conductance values come from specific conformations of the DNA and that the occurrence of these peaks can be controlled by controlling the local environment. In addition, we demonstrate that conductance measurements are uniquely sensitive to identifying these conformations in solutions and that multiple configurations can be detected in solutions over an extremely large concentration range, opening new possibilities for examining low-probability DNA conformations in solutions.

Detection and identification of genetic material via single-molecule conductance.

The ongoing discoveries of RNA modalities (for example, non-coding, micro and enhancer) have resulted in an increased desire for detecting, sequencing and identifying RNA segments for applications in food safety, water and environmental protection, plant and animal pathology, clinical diagnosis and research, and bio-security. Here, we demonstrate that single-molecule conductance techniques can be used to extract biologically relevant information from short RNA oligonucleotides, that these measurements are sensitive to attomolar target concentrations, that they are capable of being multiplexed, and that they can detect targets of interest in the presence of other, possibly interfering, RNA sequences. We also demonstrate that the charge transport properties of RNA:DNA hybrids are sensitive to single-nucleotide polymorphisms, thus enabling differentiation between specific serotypes of Escherichia coli. Using a combination of spectroscopic and computational approaches, we determine that the conductance sensitivity primarily arises from the effects that the mutations have on the conformational structure of the molecules, rather than from the direct chemical substitutions. We believe that this approach can be further developed to make an electrically based sensor for diagnostic purposes.