An official website of the United States government
Polymerase chain reaction (PCR) has long been the mainstay in genetic sequencing and identification. Irrespective of whether short read or long read technologies are adopted, PCR methods are generally time consuming and expensive. Recently, an all-electronic approach, the so-called Single Molecule Break Junction (SMBJ) method, has been proposed as a possible alternative to PCR. In this article, we evaluate the performance of four different classifier models on the current signatures of ten short strand sequences, including a pair that differs by a single mismatch. We find that a gradient boosted tree classifier model achieves impressive accuracies, ranging from approximately 96% for molecules differing by a single mismatch to 99.5% otherwise.
more »
« less
A machine learning approach for accurate and real-time DNA sequence identification
Abstract BackgroundThe 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. ResultsThis 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 %. ConclusionsA 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.
more »
« less
- PAR ID:
- 10273821
- Publisher / Repository:
- Springer Science + Business Media
- Date Published:
- Journal Name:
- BMC Genomics
- Volume:
- 22
- Issue:
- 1
- ISSN:
- 1471-2164
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Polymerase chain reaction (PCR) has long been the mainstay in genetic sequencing and identification. Irrespective of whether short read or long read technologies are adopted, PCR methods are generally time consuming and expensive. Recently, an all-electronic approach, the so-called Single Molecule Break Junction (SMBJ) method, has been proposed as a possible alternative to PCR. In this article, we evaluate the performance of four different classifier models on the current signatures of ten short strand sequences, including a pair that differs by a single mismatch. We find that a gradient boosted tree classifier model achieves impressive accuracies, ranging from approximately 96% for molecules differing by a single mismatch to 99.5% otherwise.more » « less
-
Abstract BackgroundAdding sequences into an existing (possibly user-provided) alignment has multiple applications, including updating a large alignment with new data, adding sequences into a constraint alignment constructed using biological knowledge, or computing alignments in the presence of sequence length heterogeneity. Although this is a natural problem, only a few tools have been developed to use this information with high fidelity. ResultsWe present EMMA (Extending Multiple alignments using MAFFT--add) for the problem of adding a set of unaligned sequences into a multiple sequence alignment (i.e., a constraint alignment). EMMA builds on MAFFT--add, which is also designed to add sequences into a given constraint alignment. EMMA improves on MAFFT--add methods by using a divide-and-conquer framework to scale its most accurate version, MAFFT-linsi--add, to constraint alignments with many sequences. We show that EMMA has an accuracy advantage over other techniques for adding sequences into alignments under many realistic conditions and can scale to large datasets with high accuracy (hundreds of thousands of sequences). EMMA is available athttps://github.com/c5shen/EMMA. ConclusionsEMMA is a new tool that provides high accuracy and scalability for adding sequences into an existing alignment.more » « less
-
Hancock, John (Ed.)Abstract SummaryChromosomal copy number variation (CNV) refers to a polymorphism that a DNA segment presents deletion or duplication in the population. The computational algorithms developed to identify this type of variation are usually of high computational complexity. Here we present a user-friendly R package, modSaRa, designed to perform copy number variants identification. The package is developed based on a change-point based method with optimal computational complexity and desirable accuracy. The current version of modSaRa package is a comprehensive tool with integration of preprocessing steps and main CNV calling steps. Availability and ImplementationmodSaRa is an R package written in R, C ++ and Rcpp and is now freely available for download at http://c2s2.yale.edu/software/modSaRa. Supplementary informationSupplementary data are available at Bioinformatics online.more » « less
-
Abstract MotivationTandem mass spectrometry (MS/MS) is a crucial technology for large-scale proteomic analysis. The protein database search or the spectral library search are commonly used for peptide identification from MS/MS spectra, which, however, may face challenges due to experimental variations between replicated spectra and similar fragmentation patterns among distinct peptides. To address this challenge, we present SpecEncoder, a deep metric learning approach to address these challenges by transforming MS/MS spectra into robust and sensitive embedding vectors in a latent space. The SpecEncoder model can also embed predicted MS/MS spectra of peptides, enabling a hybrid search approach that combines spectral library and protein database searches for peptide identification. ResultsWe evaluated SpecEncoder on three large human proteomics datasets, and the results showed a consistent improvement in peptide identification. For spectral library search, SpecEncoder identifies 1%–2% more unique peptides (and PSMs) than SpectraST. For protein database search, it identifies 6%–15% more unique peptides than MSGF+ enhanced by Percolator, Furthermore, SpecEncoder identified 6%–12% additional unique peptides when utilizing a combined library of experimental and predicted spectra. SpecEncoder can also identify more peptides when compared to deep-learning enhanced methods (MSFragger boosted by MSBooster). These results demonstrate SpecEncoder’s potential to enhance peptide identification for proteomic data analyses. Availability and ImplementationThe source code and scripts for SpecEncoder and peptide identification are available on GitHub at https://github.com/lkytal/SpecEncoder. Contact: hatang@iu.edu.more » « less
