Development of an Artificially Intelligent Nanopore for High-Throughput DNA Sequencing with a Machine-Learning-Aided Quantum-Tunneling Approach

Abstract

Solid-state nanopore-based single-molecule DNA sequencing with quantum tunneling technology poses formidable challenges to achieve long-read sequencing and high-throughput analysis. Here, we propose a method for developing an artificially intelligent (AI) nanopore that does not require extraction of the signature transmission function for each nucleotide of the whole DNA strand by integrating supervised machine learning (ML) and transverse quantum transport technology with a graphene nanopore. The optimized ML model can predict the transmission function of all other nucleotides after training with data sets of all the orientations of any nucleotide inside the nanopore with a rootmean-square error (RMSE) of as low as 0.062. Further, up to 96.01% accuracy is achieved in classifying the unlabeled nucleotides with their transmission readouts. We envision that an AI nanopore can alleviate the experimental challenges of the quantum-tunneling method and pave the way for rapid and high-precision DNA sequencing by predicting their signature transmission functions.

Deciphering DNA nucleotide sequences and their rotation dynamics with interpretable machine learning integrated C3N nanopores

“…The results of the N-terminated C 3 N nanopore are observed to be better than those of the H-atom-terminated graphene nanopore, where only a few rotated configurations were predicted. 28 This could be attributed to the presence of fewer outliers in the transmission datasets. Moreover, as the nanopore–nucleotide coupling and electronic properties of peripheral atoms determine the transmission of that nucleotide, cancerous epigenetic/oxidized nucleotides can also be identified by this technique.…”

“…ML classification is already reported to be efficient in classifying nucleotides with high accuracy. 28 Fig. 6a shows the workflow for ML classification.…”

“…25–27 A machine learning quantum tunneling-based AI nanopore has been also reported for single nucleotide DNA sequencing. 28…”

“…[25][26][27] A machine learning quantum tunneling-based AI nanopore has been also reported for single nucleotide DNA sequencing. 28 Recently, a new member of the graphene-like material family, two-dimensional (2D) C 3 N, has been successfully fabricated by experiment. 29,30 It is a hole-free 2D honeycomb lattice with a homogeneous and ordered distribution of carbon and nitrogen atoms.…”

Artificially Intelligent Nanogap for Rapid DNA Sequencing: A Machine Learning Aided Quantum Tunneling Approach

“…Fyta et al reported that ionic blockade height dominates over other features like traditional dwell time and mean ionic current. , Similarly, there are a few reports where ML applications have been done for DNA/amino acid sequencing with biological and solid-state nanogaps. Recently, our group has successfully established the pertinency of ML application in both DNA and protein sequencing with the transverse quantum transport method by predicting the fingerprint transmission functions. − Despite the advancement of research in the realm of single nucleotide identification and ML analysis utilizing nanogaps, a disparity still persists. Encouraged by the remarkable strides made by machine learning in the realm of sequencing, we aspired to explore the potential ML classification with a 2D material-based nanogap device for DNA sequencing application.…”