Intelligent Transportation Systems (ITS) demand real-time collision prediction to ensure road safety and reduce accident severity. Conventional approaches rely on transmitting raw video or high-dimensional sensory data from roadside units (RSUs) to vehicles, which is impractical under vehicular communication bandwidth and latency constraints. In this work, we propose a semantic V2X framework in which RSU-mounted cameras generate spatiotemporal semantic embeddings of future frames using the Video Joint Embedding Predictive Architecture (V-JEPA). To evaluate the system, we construct a digital twin of an urban traffic environment enabling the generation of d verse traffic scenarios with both safe and collision events. These embeddings of the future frame, extracted from V-JEPA, capture task-relevant traffic dynamics and are transmitted via V2X links to vehicles, where a lightweight attentive probe and classifier decode them to predict imminent collisions. By transmitting only semantic embeddings instead of raw frames, the proposed system significantly reduces communication overhead while maintaining predictive accuracy. Experimental results demonstrate that the framework with an appropriate processing method achieves a 10% F1-score improvement for collision prediction while reducing transmission requirements by four orders of magnitude compared to raw video. This validates the potential of semantic V2X communication to enable cooperative, real-time collision prediction in ITS.

We consider the design of a positioning system where a robot determines its position from local observations. This is a well-studied problem of considerable practical importance and mathematical interest. The dominant paradigm derives from the classical theory of de Bruijn sequences, where the robot has access to a window within a larger code and can determine its position if these windows are distinct. We propose an alternative model in which the robot has more limited observational powers, which we argue is more realistic in terms of engineering: the robot does not have access to the full pattern of colours (or letters) in the window, but only to the intensity of each colour (or the number of occurrences of each letter). This leads to a mathematically interesting problem with a different flavour to that arising in the classical paradigm, requiring new construction techniques. The parameters of our construction are optimal up to a constant factor, and computing the position requires only a constant number of arithmetic operations.

This paper investigates the privacy leakage of smart speakers under an encrypted traffic analysis attack, referred to as voice command fingerprinting. In this attack, an adversary can eavesdrop both outgoing and incoming encrypted voice traffic of a smart speaker, and infers which voice command a user says over encrypted traffic. We first built an automatic voice traffic collection tool and collected two large-scale datasets on two smart speakers, Amazon Echo and Google Home. Then, we implemented proof-of-concept attacks by leveraging deep learning. Our experimental results over the two datasets indicate disturbing privacy concerns. Specifically, compared to 1% accuracy with random guess, our attacks can correctly infer voice commands over encrypted traffic with 92.89% accuracy on Amazon Echo.

BACKGROUND: Comparing the composition of microbial communities among groups of interest (e.g., patients vs healthy individuals) is a central aspect in microbiome research. It typically involves sequencing, data processing, statistical analysis and graphical display. Such an analysis is normally obtained by using a set of different applications that require specific expertise for installation, data processing and in some cases, programming skills. RESULTS: Here, we present SHAMAN, an interactive web application we developed in order to facilitate the use of (i) a bioinformatic workflow for metataxonomic analysis, (ii) a reliable statistical modelling and (iii) to provide the largest panel of interactive visualizations among the applications that are currently available. SHAMAN is specifically designed for non-expert users. A strong benefit is to use an integrated version of the different analytic steps underlying a proper metagenomic analysis. The application is freely accessible at http://shaman.pasteur.fr/ , and may also work as a standalone application with a Docker container (aghozlane/shaman), conda and R. The source code is written in R and is available at https://github.com/aghozlane/shaman . Using two different datasets (a mock community sequencing and a published 16S rRNA metagenomic data), we illustrate the strengths of SHAMAN in quickly performing a complete metataxonomic analysis. CONCLUSIONS: With SHAMAN, we aim at providing the scientific community with a platform that simplifies reproducible quantitative analysis of metagenomic data.

Summary: We designed a PyQt graphical user interface-Sequanix-aimed at democratizing the use of Snakemake pipelines in the NGS space and beyond. By default, Sequanix includes Sequana NGS pipelines (Snakemake format) (http://sequana.readthedocs.io), and is also capable of loading any external Snakemake pipeline. New users can easily, visually, edit configuration files of expert-validated pipelines and can interactively execute these production-ready workflows. Sequanix will be useful to both Snakemake developers in exposing their pipelines and to a wide audience of users. Availability and implementation: Source on http://github.com/sequana/sequana, bio-containers on http://bioconda.github.io and Singularity hub (http://singularity-hub.org). Contact: [email protected] or [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.