DNA molecules can electrophoretically be driven through a nanoscale opening in a material, giving rise to rich and measurable ionic current blockades. In this work, we train machine learning models on experimental ionic blockade data from DNA nucleotide translocation through 2D pores of different diameters. The aim of the resulting classification is to enhance the read-out efficiency of the nucleotide identity providing pathways toward error-free sequencing. We propose a novel method that at the same time reduces the current traces to a few physical descriptors and trains low-complexity models, thus reducing the dimensionality of the data. We describe each translocation event by four features including the height of the ionic current blockade. Training on these lower dimensional data and utilizing deep neural networks and convolutional neural networks, we can reach a high accuracy of up to 94% in average. Compared to more complex baseline models trained on the full ionic current traces, our model outperforms. Our findings clearly reveal that the use of the ionic blockade height as a feature together with a proper combination of neural networks, feature extraction, and representation provides a strong enhancement in the detection. Our work points to a possible step toward guiding the experiments to the number of events necessary for sequencing an unknown biopolymer in view of improving the biosensitivity of novel nanopore sequencers.

2D nanopores can be used to electrophoretically drive DNA molecules, which can in turn be identified through measurable electronic current blockades. In this work, we use experimental data from molybdenum disulfide nanopores threading DNA nucleotides and propose a methodological approach to interpret DNA events. Specifically, the experimental ionic traces are used to train an unsupervised machine learning model for identifying distinct molecular events through the 2D nanopore. For the first time, we propose a clustering of experimental 2D nanopore data based on the ionic current blockade height and unrelated to the traditional dwell time for each DNA event. Within this approach, the blockade level information is implicitly included in the feature space analysis and does not need to be treated explicitly. We could show the higher efficiency of the blockade height over the traditional dwell time also in coping with sparse nanopore data sets. Our approach allows for a deep insight into characteristic molecular features in 2D nanopores and provides a feedback mechanism to tune these materials and interpret the measured signals. It has, thus, a high impact on the efficiency of 2D nanopore-based DNA sequencers.

No abstract

A very simple, fast, and efficient approach to analyze and identify respiratory-related virus sequences based on machine learning is proposed. Such schemes are very important in identifying viruses, especially in view of spreading pandemics. The method is based on genetic code rules and the open reading frame (ORF). Data from the respiratory-related coronaviruses are collected and features are extracted based on reoccurring nucleobase 3-tuples in the RNA. Our methodology is simply based on counting nucleobase triplets, normalizing the count to the length of the sequence, and applying principal component analysis (PCA) techniques. The triplet counting can be further used for classification purposes. DNA sequences from the herpes virus family can be considered as the first step towards a complete and accurate classification including more complex factors, such as mutations. The proposed classification scheme is simply based on “counting” biological information. It can serve as the first fast detection method, widely accessible and portable to a variety of distinct architectures for fast and on-the-fly detection. We provide an approach that can be further optimized and combined with supervised techniques to allow for more accurate detection and read out of the exact virus type or sequence. We discuss the relevance of this scheme in identifying differences in similar viruses and their impact on biochemical analysis.

No abstract