Deep learning in next-generation sequencing

AbstractTargeted high-throughput DNA sequencing is a primary approach for genomics and molecular diagnostics, and more recently as a readout for DNA information storage. Oligonucleotide probes used to enrich gene loci of interest have different hybridization kinetics, resulting in non-uniform coverage that increases sequencing costs and decreases sequencing sensitivities. Here, we present a deep learning model (DLM) for predicting Next-Generation Sequencing (NGS) depth from DNA probe sequences. Our DLM includes a bidirectional recurrent neural network that takes as input both DNA nucleotide identities as well as the calculated probability of the nucleotide being unpaired. We apply our DLM to three different NGS panels: a 39,145-plex panel for human single nucleotide polymorphisms (SNP), a 2000-plex panel for human long non-coding RNA (lncRNA), and a 7373-plex panel targeting non-human sequences for DNA information storage. In cross-validation, our DLM predicts sequencing depth to within a factor of 3 with 93% accuracy for the SNP panel, and 99% accuracy for the non-human panel. In independent testing, the DLM predicts the lncRNA panel with 89% accuracy when trained on the SNP panel. The same model is also effective at predicting the measured single-plex kinetic rate constants of DNA hybridization and strand displacement.

Download Full-text

Gene Prediction in Metagenomic Fragments with Deep Learning

BioMed Research International ◽

10.1155/2017/4740354 ◽

2017 ◽

Vol 2017 ◽

pp. 1-9 ◽

Cited By ~ 5

Author(s):

Shao-Wu Zhang ◽

Xiang-Yang Jin ◽

Teng Zhang

Keyword(s):

Deep Learning ◽

Next Generation Sequencing ◽

Gene Prediction ◽

Learning Model ◽

Next Generation ◽

Sequencing Technologies ◽

Novel Method ◽

Z Curve ◽

Length Coverage ◽

Generation Sequencing

Next generation sequencing technologies used in metagenomics yield numerous sequencing fragments which come from thousands of different species. Accurately identifying genes from metagenomics fragments is one of the most fundamental issues in metagenomics. In this article, by fusing multifeatures (i.e., monocodon usage, monoamino acid usage, ORF length coverage, and Z-curve features) and using deep stacking networks learning model, we present a novel method (called Meta-MFDL) to predict the metagenomic genes. The results with 10 CV and independent tests show that Meta-MFDL is a powerful tool for identifying genes from metagenomic fragments.

Download Full-text

Aprendizaje de máquina y aprendizaje profundo en biotecnología: aplicaciones, impactos y desafíos

Ciencia Ambiente y Clima ◽

10.22206/cac.2019.v2i2.pp7-26 ◽

2019 ◽

Vol 2 (2) ◽

pp. 7-26

Author(s):

Edian F. Franco ◽

Rommel J. Ramos

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Next Generation Sequencing ◽

Biological Data ◽

Next Generation ◽

Next Generation Sequencing Ngs ◽

Generation Sequencing

La bioinformática es un área que ha modificado la forma en que se diseñan y se desarrollan los experimentos e investigaciones de las áreas biológicas. La biotecnología no ha quedado fuera de los alcances de la bioinformática, impactando directamente áreas como el descubrimiento y el desarrollo de fármacos, mejoramiento de cultivos, biorremediación, estudios de la diversidad ambiental, patología molecular, entre otras. Esto se debe, en gran medida, al desarrollo de las tecnologías de secuenciación de alto rendimiento o Next-generation sequencing (NGS), que han generado gran cantidad de datos que deben ser procesados y analizados para producir nuevos conocimientos y descubrimientos. Lo anterior ha promovido que dos áreas de la bioinformática y la ciencia de la computación, machine learning y deep learning, hayan sido utilizadas para el análisis de estos datos. El “aprendizaje de máquina” aplica técnicas que permiten que las computadoras aprendan, mientras que el “aprendizaje profundo” genera modelos de redes neuronales artificiales que intenta imitar el funcionamiento del cerebro humano, permitiéndoles aprender a partir de los datos y mejorar su aprendizaje a través de las experiencias. Estas dos áreas son esenciales para poder identificar, analizar, interpretar y obtener conocimientos de la gran cantidad de datos biológicos (Big biological data). En este trabajo hacemos una revisión de estas dos áreas: el aprendizaje de máquina y el aprendizaje profundo, orientado al impacto y sus aplicaciones en el área de biotecnología.

Download Full-text

The application of deep learning for the classification of correct and incorrect SNP genotypes from whole-genome DNA sequencing pipelines

Journal of Applied Genetics ◽

10.1007/s13353-020-00586-0 ◽

2020 ◽

Vol 61 (4) ◽

pp. 607-616

Author(s):

Krzysztof Kotlarz ◽

Magda Mielczarek ◽

Tomasz Suchocki ◽

Bartosz Czech ◽

Bernt Guldbrandtsen ◽

...

Keyword(s):

Deep Learning ◽

Next Generation Sequencing ◽

Training Data ◽

Next Generation ◽

Validation Data ◽

Data Set ◽

Genotype Information ◽

Naive Algorithm ◽

Generation Sequencing

Abstract A downside of next-generation sequencing technology is the high technical error rate. We built a tool, which uses array-based genotype information to classify next-generation sequencing–based SNPs into the correct and the incorrect calls. The deep learning algorithms were implemented via Keras. Several algorithms were tested: (i) the basic, naïve algorithm, (ii) the naïve algorithm modified by pre-imposing different weights on incorrect and correct SNP class in calculating the loss metric and (iii)–(v) the naïve algorithm modified by random re-sampling (with replacement) of the incorrect SNPs to match 30%/60%/100% of the number of correct SNPs. The training data set was composed of data from three bulls and consisted of 2,227,995 correct (97.94%) and 46,920 incorrect SNPs, while the validation data set consisted of data from one bull with 749,506 correct (98.05%) and 14,908 incorrect SNPs. The results showed that for a rare event classification problem, like incorrect SNP detection in NGS data, the most parsimonious naïve model and a model with the weighting of SNP classes provided the best results for the classification of the validation data set. Both classified 19% of truly incorrect SNPs as incorrect and 99% of truly correct SNPs as correct and resulted in the F1 score of 0.21 — the highest among the compared algorithms. We conclude the basic models were less adapted to the specificity of a training data set and thus resulted in better classification of the independent, validation data set, than the other tested models.

Download Full-text