Biology subjects

Nip, K. M.

Publications and source records attributed to Nip, K. M..

9 recordsLinked to original sources

bioRxiv2024

ntEmbd: Deep learning embedding for nucleotide sequences

Enabled by the explosion of data and substantial increase in computational power, deep learning has transformed fields such as computer vision and natural language processing (NLP) and it has become a successful method to be applied to many transcriptomic analysis tasks. A core advantage of deep learning is its inherent capability to incorporate feature computation within the machine learning models. This results in a comprehensive and machine-readable representation of sequences, facilitating the downstream classification and clustering tasks. Compared to machine translation problems in NLP, feature embedding is particularly challenging for transcriptomic studies as the sequences are string of thousands of nucleotides in length, which make the long-term dependencies between features from different parts of the sequence even more difficult to capture. This highlights the need for nucleotide sequence embedding methods that are capable of learning input sequence features implicitly. Here we introduce ntEmbd, a deep learning embedding tool that captures dependencies between different features of the sequences and learns a latent representation for given nucleotide sequences. We further provide two sample use cases, describing how learned RNA features can be used in downstream analysis. The first use case demonstrates ntEmbds utility in classifying coding and noncoding RNA benchmarked against existing tools, and the second one explores the utility of learned representations in identifying adapter sequences in nanopore RNA-seq reads. The tool as well as the trained models are freely available on GitHub at https://github.com/bcgsc/ntEmbd

bioinformatics↗

bioRxiv2023

Genomic structures and regulation patterns at HPV integration sites in cervical cancer

Human papillomavirus (HPV) integration has been implicated in transforming HPV infection into cancer, but its genomic consequences have been difficult to study using short-read technologies. To resolve the dysregulation associated with HPV integration, we performed long-read sequencing on 63 cervical cancer genomes. We identified six categories of integration events based on HPV-human genomic structures. Of all HPV integrants, defined as two HPV-human breakpoints bridged by an HPV sequence, 24% contained variable copies of HPV between the breakpoints, a phenomenon we termed heterologous integration. Analysis of DNA methylation within and in proximity to the HPV genome at individual integration events revealed relationships between methylation status of the integrant and its orientation and structure. Dysregulation of the human epigenome and neighboring gene expression in cis with the HPV-integrated allele was observed over megabase-ranges of the genome. By elucidating the structural, epigenetic, and allele-specific impacts of HPV integration, we provide insight into the role of integrated HPV in cervical cancer.

genomics↗

bioRxiv2023

Systematic assessment of long-read RNA-seq methods for transcript identification and quantification

AbstractThe Long-read RNA-Seq Genome Annotation Assessment Project (LRGASP) Consortium was formed to evaluate the effectiveness of long-read approaches for transcriptome analysis. The consortium generated over 427 million long-read sequences from cDNA and direct RNA datasets, encompassing human, mouse, and manatee species, using different protocols and sequencing platforms. These data were utilized by developers to address challenges in transcript isoform detection and quantification, as well as de novo transcript isoform identification. The study revealed that libraries with longer, more accurate sequences produce more accurate transcripts than those with increased read depth, whereas greater read depth improved quantification accuracy. In well-annotated genomes, tools based on reference sequences demonstrated the best performance. When aiming to detect rare and novel transcripts or when using reference-free approaches, incorporating additional orthogonal data and replicate samples are advised. This collaborative study offers a benchmark for current practices and provides direction for future method development in transcriptome analysis.

genomics↗

bioRxiv2022

GoldRush: A de novo long read genome assembler with linear time complexity

MotivationCurrent state-of-the-art long read de novo genome assemblers follow the Overlap Layout Consensus (OLC) paradigm, an O(n2) algorithm in its naive implementation. While the most time- and memory-intensive step of OLC --the all-vs-all sequencing read alignment process-- was improved and reimplemented in modern long read assemblers, these tools still often require excessive computational memory when assembling a typical 50X human genome dataset. ResultsHere we present GoldRush, a de novo genome assembly algorithm with linear time complexity in the number of input long sequencing reads. We tested GoldRush on Oxford Nanopore Technologies datasets with different base error profiles describing the genomes of three human cell lines (NA24385, HG01243 and HG02055), Oryza sativa (rice), and Solanum lycopersicum (tomato). GoldRush achieved NGA50 lengths of 18.3-22.2 Mbp for the three human datasets, with two of the three assemblies having the fewest extensive misassemblies, and NGA50 lengths of 0.3 and 2.6 Mbp for the 373 Mbp and 824 Mbp genomes of rice and tomato, respectively. Further, GoldRush assembled all genomes within a day, using at most 54.5 GB of RAM. These results demonstrate that our algorithm and new assembly paradigm can be used to assemble large genomes de novo efficiently in compute memory space, with resulting assembly contiguity comparable to that of state-of-the-art OLC genome assemblers. Availabilityhttps://github.com/bcgsc/goldrush

genomics↗

bioRxiv2022

Reference-free assembly of long-read transcriptome sequencing data with RNA-Bloom2

Long-read sequencing technologies have improved significantly since their emergence. Their read lengths, potentially spanning entire transcripts, is advantageous for reconstructing transcriptomes. Existing long-read transcriptome assembly methods are primarily reference-based and to date, there is little focus on reference-free transcriptome assembly. We introduce RNA-Bloom2, a reference-free assembly method for long-read transcriptome sequencing data. Using simulated datasets and spike-in control data, we show that the transcriptome assembly quality of RNA-Bloom2 is competitive to those of reference-based methods. Furthermore, RNA-Bloom2 requires 27.0 to 80.6% of the peak memory and 3.6 to 10.8% of the total wall-clock runtime of a competing reference-free method. Finally, we showcase RNA-Bloom2 in assembling a transcriptome sample of Picea sitchensis (Sitka spruce). Since our method does not rely on a reference, it sets up the groundwork for large-scale comparative transcriptomics where high-quality draft genome assemblies are not readily available.

bioinformatics↗

bioRxiv2021

Characterization and simulation of metagenomic nanopore sequencing data with Meta-NanoSim

Nanopore sequencing is crucial to metagenomic studies as its kilobase-long reads can contribute to resolving genomic structural differences among microbes. However, platform-specific challenges, including high base-call error rate, non-uniform read lengths, and the presence of chimeric artifacts, necessitate specifically designed analytical tools. Here, we present Meta-NanoSim, a fast and versatile utility that characterizes and simulates the unique properties of nanopore metagenomic reads. Further, Meta-NanoSim improves upon state-of-the-art methods on microbial abundance estimation through a base-level quantification algorithm. We demonstrate that Meta-NanoSim simulated data can facilitate the development of metagenomic algorithms and guide experimental design through a metagenomic assembly benchmarking task.

bioinformatics↗

bioRxiv2019

Trans-NanoSim characterizes and simulates nanopore RNA-seq data

We introduce Trans-NanoSim, the first tool that simulates reads with technical and transcriptome-specific features learnt from nanopore RNA-seq data. Through benchmarking on sets of nanopore reads from human and mouse reference transcriptomes, we show the robustness of Trans-NanoSim in capturing the characteristics of nanopore cDNA and direct RNA reads. As a cost-effective alternative to sequencing real transcriptomes, Trans-NanoSim would facilitate the rapid development of analytical tools for nanopore RNA-seq data. Trans-NanoSim is freely accessible at https://github.com/bcgsc/NanoSim

bioinformatics↗

bioRxiv2019

Termin(A)ntor: Polyadenylation Site Prediction using deep learning models

As a widespread RNA processing machinery, alternative polyadenylation plays a crucial role in gene regulation. To help decipher its underlying mechanism and understand its impact, it is desirable to comprehensively profile 3-untranslated region cleavage and associated polyadenylation sites. State-of-the-art polyadenylation site detection tools are known to be influenced by library preparation artefacts or manually selected features. Moreover, recently published machine learning methods have only been tested on pre-constructed datasets, thus lacking validation on experimental data. Here we present Terminitor, the first deep neural network-based profiling pipeline to make predictions from RNA-seq data. We show how Terminitor outperforms competing tools in sensitivity and precision on experimental transcriptome sequencing data, and demonstrate its use with data from short- and long-read sequencing technologies. For species without a good reference transcriptome annotation, Terminitor is still able to pass on the information learnt from a related species and make reasonable predictions. We used Terminitor to showcase how single nucleotide variations can create or destroy polyadenylated cleavage sites in human RNA-seq samples. Author Summary3 cleavage and polyadenylation of pre-mRNA is part of RNA maturation process. One gene can be cleaved at different positions at its 3 end, namely alternatively polyadenylation, thus identifying the correct polyadenylated cleavage site (poly(A) CS) is essential to unveil its role in gene regulation under different physiological and pathological conditions. The current poly(A) CS prediction tools are either heavily influenced by RNA-Seq library preparation artefacts or have only been designed and tested on ad hoc datasets, lacking association with real world applications. In this study, we present a deep learning model, Terminitor, that predicts the probability of a nucleotide sequence containing a poly(A) CS, and validated its performance on human and mouse data. Along with the model, we propose a poly(A) CS profiling pipeline for RNA-seq data. We benchmarked our pipeline against competing tools and achieved higher sensitivity and precision in experimental data. The usage of Terminitor is not limited to genome and transcriptome annotation and we expect it to facilitate the identification of novel isoforms, improve the accuracy of transcript quantification and differential expression analysis, and contribute to the repertoire of reference transcriptome annotation.

bioinformatics↗

bioRxiv2019

RNA-Bloom provides lightweight reference-free transcriptome assembly for single cells

We present RNA-Bloom, a de novo RNA-seq assembly algorithm that leverages the rich information content in single-cell transcriptome sequencing (scRNA-seq) data to reconstruct cell-specific isoforms. We benchmark RNA-Blooms performance against leading bulk RNA-seq assembly approaches, and illustrate its utility in detecting cell-specific gene fusion events using sequencing data from HiSeq-4000 and BGISEQ-500 platforms. We expect RNA-Bloom to boost the utility of scRNA-seq data, expanding what is informatically accessible now.

bioinformatics↗