Dataset Information

Deep learning and direct sequencing of labeled RNA captures transcriptome dynamics.

ABSTRACT: Quantification of the dynamics of RNA metabolism is essential for understanding gene regulation in health and disease. Existing methods rely on metabolic labeling of nascent RNAs and physical separation or inference of labeling through PCR-generated mutations, followed by short-read sequencing. However, these methods are limited in their ability to identify transient decay intermediates or co-analyze RNA decay with cis-regulatory elements of RNA stability such as poly(A) tail length and modification status, at single molecule resolution. Here we use 5-ethynyl uridine (5EU) to label nascent RNA followed by direct RNA sequencing with nanopores. We developed RNAkinet, a deep convolutional and recurrent neural network that processes the electrical signal produced by nanopore sequencing to identify 5EU-labeled nascent RNA molecules. RNAkinet demonstrates generalizability to distinct cell types and organisms and reproducibly quantifies RNA kinetic parameters allowing the combined interrogation of RNA metabolism and cis-acting RNA regulatory elements.

SUBMITTER: Martinek V

PROVIDER: S-EPMC10680836 | biostudies-literature | 2023 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep learning and direct sequencing of labeled RNA captures transcriptome dynamics.

Martinek Vlastimil V Martin Jessica J Belair Cedric C Payea Matthew J MJ Malla Sulochan S Alexiou Panagiotis P Maragkakis Manolis M

bioRxiv : the preprint server for biology 20231117

Quantification of the dynamics of RNA metabolism is essential for understanding gene regulation in health and disease. Existing methods rely on metabolic labeling of nascent RNAs and physical separation or inference of labeling through PCR-generated mutations, followed by short-read sequencing. However, these methods are limited in their ability to identify transient decay intermediates or co-analyze RNA decay with cis-regulatory elements of RNA stability such as poly(A) tail length and modifica ...[more]

PMID: 38014155

Similar Datasets

Project description:Conventional bacterial genome annotation provides information about coding sequences but ignores untranslated regions and operons. However, untranslated regions contain important regulatory elements as well as targets for many regulatory factors, such as small RNAs. Operon maps are also essential for functional gene analysis. In the last decade, considerable progress has been made in the study of bacterial transcriptomes through transcriptome sequencing (RNA-seq). Given the compact nature of bacterial genomes, many challenges still cannot be resolved through short reads generated using classical RNA-seq because of fragmentation and loss of the full-length information. Direct RNA sequencing is a technology that sequences the native RNA directly without information loss or bias. Here, we employed direct RNA sequencing to annotate the Vibrio parahaemolyticus transcriptome with its full features, including transcription start sites (TSSs), transcription termination sites, and operon maps. A total of 4,103 TSSs were identified. In comparison to short-read sequencing, full-length information provided a deeper view of TSS classification, showing that most internal and antisense TSSs were actually a result of gene overlap. Sequencing the transcriptome of V. parahaemolyticus grown with bile allowed us to study the landscape of pathogenicity island Vp-PAI. Some genes in this region were reannotated, providing more accurate annotation to increase precision in their characterization. Quantitative detection of operons in V. parahaemolyticus showed high complexity in some operons, shedding light on a greater extent of regulation within the same operon. Our study using direct RNA sequencing provides a quantitative and high-resolution landscape of the V. parahaemolyticus transcriptome. IMPORTANCE Vibrio parahaemolyticus is a halophilic bacterium found in the marine environment. Outbreaks of gastroenteritis resulting from seafood poisoning by these pathogens have risen over the past 2 decades. Upon ingestion by humans-often through the consumption of raw or undercooked seafood-V. parahaemolyticus senses the host environment and expresses numerous genes, the products of which synergize to synthesize and secrete toxins that can cause acute gastroenteritis. To understand the regulation of such adaptive response, mRNA transcripts must be mapped accurately. However, due to the limitations of common sequencing methods, not all features of bacterial transcriptomes are always reported. We applied direct RNA sequencing to analyze the V. parahaemolyticus transcriptome. Mapping the full features of the transcriptome is anticipated to enhance our understanding of gene regulation in this bacterium and provides a data set for future work. Additionally, this study reveals a deeper view of a complicated transcriptome landscape, demonstrating the importance of applying such methods to other bacterial models.

Project description:BackgroundPolygenic risk score (PRS) prediction is widely used to assess the risk of diagnosis and progression of many diseases. Routinely, the weights of individual SNPs are estimated by the linear regression model that assumes independent and linear contribution of each SNP to the phenotype. However, for complex multifactorial diseases such as Alzheimer's disease, diabetes, cardiovascular disease, cancer, and others, association between individual SNPs and disease could be non-linear due to epistatic interactions. The aim of the presented study is to explore the power of non-linear machine learning algorithms and deep learning models to predict the risk of multifactorial diseases with epistasis.MethodsSimulated data with 2- and 3-loci interactions and tested three different models of epistasis: additive, multiplicative and threshold, were generated using the GAMETES. Penetrance tables were generated using PyTOXO package. For machine learning methods we used multilayer perceptron (MLP), convolutional neural network (CNN) and recurrent neural network (RNN), Lasso regression, random forest and gradient boosting models. Performance of machine learning models were assessed using accuracy, AUC-ROC, AUC-PR, recall, precision, and F1 score.ResultsFirst, we tested ensemble tree methods and deep learning neural networks against LASSO linear regression model on simulated data with different types and strength of epistasis. The results showed that with the increase of strength of epistasis effect, non-linear models significantly outperform linear. Then the higher performance of non-linear models over linear was confirmed on real genetic data for multifactorial phenotypes such as obesity, type 1 diabetes, and psoriasis. From non-linear models, gradient boosting appeared to be the best model in obesity and psoriasis while deep learning methods significantly outperform linear approaches in type 1 diabetes.ConclusionOverall, our study underscores the efficacy of non-linear models and deep learning approaches in more accurately accounting for the effects of epistasis in simulations with specific configurations and in the context of certain diseases.

Dataset Information

Deep learning and direct sequencing of labeled RNA captures transcriptome dynamics.

Publications

Deep learning and direct sequencing of labeled RNA captures transcriptome dynamics.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets