Quantitative measurement of RNA expression levels through RNA-Seq is an ideal replacement for conventional cancer diagnosis via microscope examination. Currently, cancer-related RNA-Seq studies focus on two aspects: classifying the status and tissue of origin of a sample and discovering marker genes. Existing studies typically identify marker genes by statistically comparing healthy and cancer samples. However, this approach overlooks marker genes with low expression level differences and may be influenced by experimental results. This paper introduces "GENESO," a novel framework for pan-cancer classification and marker gene discovery using the occlusion method in conjunction with deep learning. we first trained a baseline deep LSTM neural network capable of distinguishing the origins and statuses of samples utilizing RNA-Seq data. Then, we propose a novel marker gene discovery method called "Symmetrical Occlusion (SO)". It collaborates with the baseline LSTM network, mimicking the "gain of function" and "loss of function" of genes to evaluate their importance in pan-cancer classification quantitatively. By identifying the genes of utmost importance, we then isolate them to train new neural networks, resulting in higher-performance LSTM models that utilize only a reduced set of highly relevant genes. The baseline neural network achieves an impressive validation accuracy of 96.59% in pan-cancer classification. With the help of SO, the accuracy of the second network reaches 98.30%, while using 67% fewer genes. Notably, our method excels in identifying marker genes that are not differentially expressed. Moreover, we assessed the feasibility of our method using single-cell RNA-Seq data, employing known marker genes as a validation test.

Long non-coding RNAs (lncRNAs) are key regulators of major biological processes and their functional modes are dictated by their subcellular localization. Relative nuclear enrichment of lncRNAs compared to mRNAs is a prevalent phenomenon but the molecular mechanisms governing their nuclear retention in cells remain largely unknown. Here in this study, we harness the recently released eCLIP data for a large number of RNA-binding proteins (RBPs) in K562 and HepG2 cells and utilize multiple bioinformatics methods to comprehensively survey the roles of RBPs in lncRNA nuclear retention. We identify an array of splicing RBPs that bind to nuclear-enriched lincRNAs (large intergenic non-coding RNAs) thus may act as trans-factors regulating their nuclear retention. Further analyses reveal that these RBPs may bind with distinct core motifs, flanking sequence compositions, or secondary structures to drive lincRNA nuclear retention. Moreover, network analyses uncover potential co-regulatory RBP clusters and the physical interaction between HNRNPU and SAFB2 proteins in K562 cells is further experimentally verified. Altogether, our analyses reveal previously unknown factors and mechanisms that govern lincRNA nuclear localization in cells.

Recent advances in transcriptomics have uncovered lots of novel transcripts in plants. To annotate such transcripts, dissecting their coding potential is a critical step. Computational approaches have been proven fruitful in this task; however, most current tools are designed/optimized for mammals and only a few of them have been tested on a limited number of plant species. In this work, we present NAMS webserver, which contains a novel coding potential classifier, NAMS, specifically optimized for plants. We have evaluated the performance of NAMS using a comprehensive dataset containing more than 3 million transcripts from various plant species, where NAMS demonstrates high accuracy and remarkable performance improvements over state-of-the-art software. Moreover, our webserver also furnishes functional annotations, aiming to provide users informative clues to the functions of their transcripts. Considering that most plant species are poorly characterized, our NAMS webserver could serve as a valuable resource to facilitate the transcriptomic studies. The webserver with testing dataset is freely available at http://sunlab.cpy.cuhk.edu.hk/NAMS/.

It is well known that functional modes of lncRNAs are intimately associated with their subcellular localization1and relative nuclear enrichment of lncRNAs compared to mRNAs is a prevalent phenomenon2. As RBPs control the production, maturation, localization, translation, and degradation of cellular RNAs3, we reason that it is important to uncover partner RBPs that can bind and facilitate lncRNA nuclear localization. In this study, we thus harnessed the recently released large scale of eCLIP data4and subcellular RNA-seq data5available in K562 and HepG2 cell lines to characterize lncRNA-RBP interactome and uncovered potential factors and associated mechanisms determining lncRNA nuclear retention. Analyses of the subcellular RNA-seq data identified nuclear enriched lncRNAs (nuc-lncRNAs) and confirmed that lncRNAs and eRNAs (enhancer associated lncRNAs) are relatively nuclear enriched in both cells. By integrating the RBP binding profiles, we next generated RBP/nuc-lncRNA interaction map to identify RBPs associated with nuc-lncRNAs including HNRNPU, SAFB2, KHSRP and KHDRBS1 in K562 and HRNPNPC as well as HNRNPL in HepG2 cell lines. To further confirm the above findings, HNRNPU was knocked down and which led to nuclear retention of a panel of lncRNAs.

MOTIVATION: Skeletal muscles have indispensable functions and also possess prominent regenerative ability. The rapid emergence of Next Generation Sequencing (NGS) data in recent years offers us an unprecedented perspective to understand gene regulatory networks governing skeletal muscle development and regeneration. However, the data from public NGS database are often in raw data format or processed with different procedures, causing obstacles to make full use of them. RESULTS: We provide SKmDB, an integrated database of NGS information in skeletal muscle. SKmDB not only includes all NGS datasets available in the human and mouse skeletal muscle tissues and cells, but also provide preliminary data analyses including gene/isoform expression levels, gene co-expression subnetworks, as well as assembly of putative lincRNAs, typical and super enhancers and transcription factor hotspots. Users can efficiently search, browse and visualize the information with the well-designed user interface and server side. SKmDB thus will offer wet lab biologists useful information to study gene regulatory mechanisms in the field of skeletal muscle development and regeneration. AVAILABILITY AND IMPLEMENTATION: Freely available on the web at http://sunlab.cpy.cuhk.edu.hk/SKmDB. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Motivation: Tissue biopsy is commonly used in cancer diagnosis and molecular studies. However, advanced skills are required for determining cancerous status of biopsies and tissue origin of tumor for cancerous ones. Correct classification is essential for downstream experiment design and result interpretation, especially in molecular cancer studies. Methods for accurate classification of cancerous status and tissue origin for pan-cancer biopsies are thus urgently needed. Results: We developed a deep learning-based classifier, named GeneCT, for predicting cancerous status and tissue origin of pan-cancer biopsies. GeneCT showed high performance on pan-cancer datasets from various sources and outperformed existing tools. We believe that GeneCT can potentially facilitate cancer diagnosis, tumor origin determination and molecular cancer studies. Availability and implementation: GeneCT is implemented in Perl/R and supported on GNU/Linux platforms. Source code, testing data and webserver are freely available at http://sunlab.cpy.cuhk.edu.hk/GeneCT/. Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: Thousands of long noncoding RNAs (lncRNAs) were newly identified from high throughput RNA-seq data. Functional annotation and prioritization of these lncRNAs for further experimental validation as well as the functional investigation is the bottleneck step for many noncoding RNA studies. Results: Here we describe lncFunTK that can run either as standard application or webserver for this purpose. It integrates high throughput sequencing data (i.e. ChIP-seq, CLIP-seq and RNA-seq) to construct the regulatory network associated with lncRNAs. Through the network, it calculates the Functional Information Score (FIS) of each individual lncRNA for prioritizing and inferring its functions through Gene Ontology (GO) terms of neighboring genes. In addition, it also provides utility scripts to support the input data preprocessing and the parameter optimizing. We further demonstrate that lncFunTK can be widely used in various biological systems for lncRNA prioritization and functional annotation. Availability and implementation: The lncFunTK standalone version is an open source package and freely available at http://sunlab.cpy.cuhk.edu.hk/lncfuntk under the MIT license. A webserver implementation is also available at http://sunlab.cpy.cuhk.edu.hk/lncfuntk/runlncfuntk.html. Supplementary information: Supplementary data are available at Bioinformatics online.