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

Abstract

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.

High conopeptide diversity in Conus striatus: Revealed by integration of two transcriptome sequencing platforms

“…According to the existence of 5'-end primer, 3'-end primer, or polyA tail, these sequences were divided into FLNC reads and non-full-length sequences. The former was clustered by iterative clustering in Iterative Clustering for Error Correction (Wang et al, 2021) algorithm software to generate the cluster consensus sequences. Subsequently, we corrected the polished consensus sequences of the TGS data through LoRDEC software (PacBio) with default parameters, and any redundancy in corrected consensus reads was removed by CD-hit (version 4.7) (Li and Godzik, 2006) to obtain the fi nal transcripts for subsequent analysis.…”

“…The TGS technology is also called the single-molecule real-time (SMRT) sequencing technology, which is developed by Pacific Biosciences (PacBio) with longer sequencing lengths, full-length transcripts, direct sequencing without fragmentation or post-sequencing assembly, and easy analysis of alternative splicing (AS) (Kuang et al, 2019). The TGS can provide not only nucleotide sequences of the target molecules but also information regarding epigenetic modifications for systematical investigations of gene expression in various Conus species (Levene et al, 2003;Wang et al, 2021).…”

Identifying SARS-CoV-2 Variants using Single-Molecule Conductance Measurements

“…Thus, to differentiate between all possible targets including the complementary and mismatched ones, and to automate and speed up the diagnosis process, we apply a machine learning algorithm previously developed by our group based on the XGBoost algorithm [49][50][51] . This approach decreases the number of conductance measurements required while improving the detection and differentiation accuracy.…”

Machine learning and analytical methods for single-molecule conductance measurements

“…In supervised learning, the model is provided with both the training data and the correct answer, and it uses this information to predict the objective variable for unknown data. Regression, [132][133][134][135] which predicts continuous values, and classification, 33,[136][137][138][139][140][141][142][143][144][145][146][147] which predicts categorical values, such as chemical species prediction, are included in supervised learning.…”