Network security has been a significant concern in recent years. Due to the rising number and complexity of cyber-attacks, Machine Learning (ML) models have been proposed to enhance intrusion detection. However, training these models requires extensive data, which is challenging to collect. To tackle this issue, previous work has focused on the generation of synthetic network traffic data using generative neural networks. Considering network traffic as a sequence of packets that contains continuous, discrete, and binary features, we propose a novel approach to learn a discrete representation of network traffic using Vector-Quantized Variational Autoencoders (VQ-VAE). In this paper, we adapt this model to learn how to represent network flows as a sequence of discrete tokens, called NetGlyphs. We evaluate the model on a dataset of Command & Control flows and compare performances to another model that uses a continuous representation. We show that our model is able to, reconstruct the data accurately and better preserve the original distribution. We also find promising results on network traffic generation using a state-of-the-art Transformer model to generate new NetGlyphs sequences that can be decoded back into real network traffic.

Machine learning (ML) is a promising technology for network intrusion detection systems. There is a wide range of ML algorithms that are potential candidates for network intrusion detection systems, as they exhibit very good detection accuracy in average. However, significant detection differences appear when facing different kinds of attacks, some being prone to better detect some particular attack types. They then often appear to complement each other. The challenge then lies in determining the accurate result when several ML models provide different results, and this without any explanation about their decision. To address this challenge, our system aims to reconstruct attack patterns from the outputs of these ML models and presenting them in an interpretable manner. For that, we propose an approach combining ensemble learning and stacking with a meta-learner that works on graphical representation of traffic flows, that then provides the required explainability level for the decisions made. The evaluation of our system, using the CSE-CIC-IDS2018 dataset, demonstrates a significant improvement achieved through the combination of multiple ML algorithms. Furthermore, we emphasize the importance of explainability in network intrusion detection systems and the need for accurate and interpretable models. Our system goes beyond traditional detection methods by reporting anomalous feature pairs and providing visual representations of attack patterns, empowering analysts to better understand and respond to network threats.

MOTIVATION: Long-read sequencing technologies are invaluable for determining complex RNA transcript architectures but are error-prone. Numerous 'hybrid correction' algorithms have been developed for genomic data that correct long reads by exploiting the accuracy and depth of short reads sequenced from the same sample. These algorithms are not suited for correcting more complex transcriptome sequencing data. RESULTS: We have created a novel reference-free algorithm called Transcript-level Aware Long-Read Correction (TALC) which models changes in RNA expression and isoform representation in a weighted De Bruijn graph to correct long reads from transcriptome studies. We show that transcript-level aware correction by TALC improves the accuracy of the whole spectrum of downstream RNA-seq applications and is thus necessary for transcriptome analyses that use long read technology. AVAILABILITY AND IMPLEMENTATION: TALC is implemented in C++ and available at https://github.com/lbroseus/TALC. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract Result: We have developed miREval 2.0, an online tool that can simultaneously search up to 100 sequences for novel microRNAs (miRNAs) in multiple organisms. miREval 2.0 uses multiple published in silico approaches to detect miRNAs in sequences of interest. This tool can be used to discover miRNAs from DNA sequences or to validate candidates from sequencing data. Availability: http://mimirna.centenary.org.au/mireval/. Contact: [email protected] Supplementary information: Supplementary data are available at Bioinformatics online.

Abstract Motivation: microRNAs are short non-coding RNAs that regulate gene expression by inhibiting target mRNA genes. Next-generation sequencing combined with bioinformatics analyses provide an opportunity to predict numerous novel miRNAs. The efficiency of these predictions relies on the set of positive and negative controls used. We demonstrate that commonly used positive and negative controls may be unreliable and provide a rational methodology with which to replace them. Contact: [email protected] Supplementary information: Supplementary data are available at Bioinformatics online.

MOTIVATION: microRNAs (miRNAs) are short non-coding RNAs that regulate gene expression by inhibiting target mRNA genes. Their tissue- and disease-specific expression patterns have immense therapeutic and diagnostic potential. To understand these patterns, a reliable compilation of miRNA and mRNA expression data is required to compare multiple tissue types. Moreover, with the appropriate statistical tools, such a resource could be interrogated to discover functionally related miRNA-mRNA pairs. RESULTS: We have developed mimiRNA, an online resource that integrates expression data from 1483 samples and permits visualization of the expression of 635 human miRNAs across 188 different tissues or cell types. mimiRNA incorporates a novel sample classification algorithm, ExParser, that groups identical miRNA or mRNA experiments from separate sources. This enables mimiRNA to provide reliable expression profiles and to discover functional relations between miRNAs and mRNAs such as miRNA targets. Additionally, mimiRNA incorporates a decision tree algorithm to discover distinguishing miRNA features between two tissue or cell types. We validate the efficacy of our resource on independent experimental data and through biologically relevant analyses. AVAILABILITY: http://mimirna.centenary.org.au. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

microRNAs (miRNAs) are major regulators of gene expression and thereby modulate many biological processes. Computational methods have been instrumental in understanding how miRNAs bind to mRNAs to induce their repression but have proven inaccurate. Here we describe a novel method that combines expression data from human and mouse to discover conserved patterns of expression between orthologous miRNAs and mRNA genes. This method allowed us to predict thousands of putative miRNA targets. Using the luciferase reporter assay, we confirmed 4 out of 6 of our predictions. In addition, this method predicted many miRNAs that act as expression enhancers. We show that many miRNA enhancer effects are mediated through the repression of negative transcriptional regulators and that this effect could be as common as the widely reported repression activity of miRNAs. Our findings suggest that the indirect enhancement of gene expression by miRNAs could be an important component of miRNA regulation that has been widely neglected to date.

Abstract Summary: We have developed an online tool called mirEval which can search sequences of up to 10 000 nt for novel microRNAs in multiple organisms. It is a comprehensive tool, easy to use and very informative. It will allow users with no prior knowledge of in-silico detection of microRNAs to take advantage of the most successful approaches to investigate sequences of interest. Availability: The mirEval web server is available at http://tagc.univ-mrs.fr/mireval Contact: [email protected] Supplementary information: Supplementary data are available at Bioinformatics online.

Most mammalian genes are able to express several splice variants in a phenomenon known as alternative splicing. Serious alterations of alternative splicing occur in cancer tissues, leading to expression of multiple aberrant splice forms. Most studies of alternative splicing defects have focused on the identification of cancer-specific splice variants as potential therapeutic targets. Here, we examine instead the bulk of non-specific transcript isoforms and analyze their level of disorder using a measure of uncertainty called Shannon's entropy. We compare isoform expression entropy in normal and cancer tissues from the same anatomical site for different classes of transcript variations: alternative splicing, polyadenylation, and transcription initiation. Whereas alternative initiation and polyadenylation show no significant gain or loss of entropy between normal and cancer tissues, alternative splicing shows highly significant entropy gains for 13 of the 27 cancers studied. This entropy gain is characterized by a flattening in the expression profile of normal isoforms and is correlated to the level of estimated cellular proliferation in the cancer tissue. Interestingly, the genes that present the highest entropy gain are enriched in splicing factors. We provide here the first quantitative estimate of splicing disruption in cancer. The expression of normal splice variants is widely and significantly disrupted in at least half of the cancers studied. We postulate that such splicing disorders may develop in part from splicing alteration in key splice factors, which in turn significantly impact multiple target genes.

Alternative polyadenylation sites produce transcript isoforms with 3' untranslated regions (UTRs) of different lengths. If a microRNA (miRNA) target is present in the UTR, then only those target-containing isoforms should be sensitive to control by a cognate miRNA. We carried out a systematic examination of 3' UTRs containing multiple poly(A) sites and putative miRNA targets. Based on expressed sequence tag (EST) counts and EST library information, we observed that levels of isoforms containing targets for miR-1 or miR-124, two miRNAs causing downregulation of transcript levels, were reduced in tissues expressing the corresponding miRNA. This analysis was repeated for all conserved 7-mers in 3' UTRs, resulting in a selection of 312 motifs. We show that this set is significantly enriched in known miRNA targets and mRNA-destabilizing elements, which validates our initial hypothesis. We scanned the human genome for possible cognate miRNAs and identified phylogenetically conserved precursors matching our motifs. This analysis can help identify target-miRNA couples that went undetected in previous screens, but it may also reveal targets for other types of regulatory factors.