Evaluating wildfire risk systems using standard machine-learning metrics such as F1-score or IoU is fundamentally flawed: these metrics assess event prediction accuracy, not the operational coherence of a continuous risk signal. This work proposes a novel monotonic evaluation framework that measures whether increases in a predicted risk score consistently correspond to increases in observed operational load, such as number of fires, intervention time, and deployed resources. Moreover, we compare three structurally different approaches on the French Alpes-Maritimes department: the expert-based DFE index, GRU- based predictive models, and FARS, a hybrid multi-agent system combining predictive AI with LLM-based reasoning. Experimental results reveal that the DFE, despite poor classification metrics, exhibits the most balanced monotonic behavior across the full risk scale. GRU models achieve strong local monotonicity but fail to produce well-distributed risk levels. FARS inherits and reveals the structural limitations of upstream signals rather than correcting them. The central finding is a paradigm shift: a good risk model does not predict fires accurately, but one whose ordinal scale meaningfully explains operational dynamics, as proved in this paper. Code of the monotonic framework is available on github.

Antibiotic resistance in Mycobacterium tuberculosis remains a critical global health challenge, demanding harmonized, large-scale data to identify and predict resistanceconferring mutations. We present a unified, mutation-level database of clinical M. tuberculosis isolates linking whole-genome variation to drug resistance phenotypes across a broad spectrum of first- and second-line antibiotics. The resource integrates: (i) curated repositories (including CRyPTIC and a Nature study) encompassing approximately 12,000 genomes with phenotypes [1]; (ii) an automated large language model (LLM) text-mining pipeline that scanned over 20,000 articles to extract SRAlinked resistance/susceptibility statements (approximately $\mathbf{2, 0 0 0}$ additional entries) [2], [3]; and (iii) the PATRIC compendium mapped to SRA identifiers, adding thousands of strains with annotated drug responses [4]. After reconciling strain names with NCBI SRA accessions, we standardize drug nomenclature, deduplicate records, and produce a genome-wide, binary mutation matrix (over 150,000 features per isolate) paired with resistance/susceptibility labels. This database consolidates established and newly reported mutations, expands coverage for recently introduced or repurposed antibiotics, and is immediately ML-ready. Preliminary models (XGBoost) achieve strong performance and reproduce known drivers (e.g., rpoB, katG, gyrA) while surfacing potential multimutation interactions [5], [6]. As a standardized, richly annotated training set, this resource enables robust AI development and benchmarking for resistance prediction, supports genomic surveillance, and accelerates diagnostic and therapeutic design in TB.

Forest-fire prediction using Artificial Intelligence (AI) continues to face major challenges, including (i) the ability to generalize across regions with very different risk profiles, (ii) managing the inherent daily variability and randomness of fire occurrences (including extreme fire days). These factors together have hindered the deployment of dependable prediction systems in operational settings. In this work, we introduce a novel multi-risk modeling framework specifically designed to tackle all two challenges simultaneously. The proposed approach is applied to daily forest-fire prediction across mainland France. We develop a voting-based system that combines the outputs of multiple models trained on signals smoothed with a range of convolutional kernels, capturing both local and seasonal variations. The proposed solution achieves superior performance compared to conventional models, demonstrating improved cross-regional transferability and robustness to daily fluctuations. Notably, it significantly enhances prediction skill for the rare but damaging extreme-fire days, where traditional models often fail. Our experiments reveal that using an ensemble of multiple risk models can better capture the complex dynamics of fire risk and provide more reliable guidance for decision-makers. Supplementary materials are available here.

The ever-growing use of digital systems has led to the accumulation of vast datasets, particularly time series, depicting the temporal evolution of variables and systems.Analysing these time series presents a tremendous challenge due to their inherent complexity and heterogeneity.Addressing an industrial need in the pharmaceutical wholesale sector, this paper introduces a new clustering method for time series: TSCAPE.The TSCAPE method uses a distance matrix calculated using dynamic time warping, followed by multidimensional scaling to project time series into a 2D Euclidean space, thus improving the last clustering stage by K-Means.Unlike conventional techniques, this approach, based on clustering of representation of distances in a Euclidean plane rather than on the curve shape, directly enhances the efficiency of the clustering process.The methodology exhibits significant potential for diverse applications, accommodating varied data types and irregular time series shapes.The research compares multiple variants and proposes metrics to assess their effectiveness on two open-access datasets.The results demonstrate the method's superiority over "only distance comparison clustering techniques", like dynamic time warping and K-Means, with future prospects aimed at predictive applications and refining the clustering process by exploring alternative, more powerful clustering algorithms.

International audience

Fire brigades’ operations vary with time, climate, season, occasions, etc. For example, the frequency of accidents is greater during the day than at night. Thus, adjusting the need to the demand of fire departments by categories of operations can lead to a reduction of material, financial and human resources. Therefore, it can be very helpful during the financial and economic crisis most countries face. It also helps firefighters to be well prepared by knowing the type and number of human resources needed for the next operation. The aim of this study is to predict the number of firefighters’ interventions of 14 different categories varying between emergency and non-emergency deployments. The experiments in this study on the dataset provided by the fire and rescue service, SDIS 25, in the Doubs-France region showed that it is not necessary to improve the prediction when more explanatory variables are added. Some characteristics are not informative and may reduce the accuracy of the results.

Saving a few minutes in the rescue of a person having an infactus or for a drowning can save lives. The means to save this time are therefore actively sought by the teams in charge of emergency rescue, such as the fire department in France. Part of the answer lies in the fact that some of these accidents have a predictive character: people swim outdoors especially in summer, and when it is hot and the air is polluted, there is an increased risk of discomfort and respiratory distress. The aim of this article is to describe how to experimentally implement an operational predictive tool for emergency rescue. We will study various models, and present the results obtained by the best of them, whose scores effectively allow the solution to be considered operational.

Forecasting the number of fire department deployments for different types of operations is important to size the need to the demand and hence, improve emergency response efficiency and reduce financial and material resources. Fire department operations are not considered hazardous because they are somewhat related to time and date. Fires are more likely to occur in the fall than in the winter, and floods are more risky in the winter than in the spring. Car accidents are also logically more likely to occur during the day than at night, when most people are resting at home. This work focuses on predicting the target value of fire calls by creating 14 different subsets of data for each type of possible category (childbirth, fire, suicide, traffic accident, drown, fire on public road, water-flood, heating, emergency aid to people, help for people, public road accident, brawl, witness, and wasp). The methodology was based on the Departmental Fire and Rescue Doubs (SDIS 25) in France, where two machine learning techniques were then implemented to verify the feasibility of the experiments. Although the results can be improved by adding additional explanatory variables, the results were promising.

Most research on energy efficiency in wireless sensor networks considers that the communication subsystem consumes significantly more energy than the sensing and computing ones. In order to verify this widely adopted premise, an experimental study has been conducted on a Pysense sensor shield that utilizes WiFi and LoRa. This paper compares the energy consumption of each subsystem and two data-driven energy conservation algorithms that employ different strategies. The findings of this work indicate that lowering the energy consumption of the communication subsystem is only advantageous when using WiFi but was less effective and promising when using LoRa. Additionally, it demonstrates the importance of simultaneously optimizing the activation of multiple subsystems to minimize energy consumption. The findings of this study, as well as the source code, are available on Github: https://github.com/BouTayehGaby/WSN-energy-consumption-benchmark.

Emergency medical services (EMS) provide crucial prehospital care, such as in the case of cardiac arrest, where the victim requires immediate first-aid. For this reason, it is vital to improving EMS response time. This article proposes a novel methodology based on machine learning (ML) techniques to predict both the victims’ mortality and their need for transportation to health facilities using data gathered from the start of the emergency call until the Departmental Fire and Rescue Service of the Doubs (SDIS25) is notified. We first analyzed SDIS25 calls to find out associations between the call processing times and victims’ mortality, and to measure the variables’ importance. Next, we validated our proposed ML-based methodology, where mortality could be predicted with accuracy and area under the receiver operating characteristic curve (AUC) scores of 96.44% and 96.04%, respectively, while the need for transportation achieved an accuracy and AUC scores of 73.62% and 78.91%, respectively. What is more, we found out that it was still possible to predict both targets perturbating the input data by applyingk-anonymity and differential privacy techniques. In conclusion, the results showed the potential of ML for EMS, which can be used as a decision-support tool to early identify mortality and the use of resources (transportation) and, thus, help EMS to save more lives and avoid service disruptions.

Firefighters are equipped with an immobility detector device also called the Personal Alert Safety System (PASS) that is integrated into the user's Self-Contained Breathing Apparatus (SCBA). If a firefighter remains motionless for a certain period of time, a loud audible alert is triggered to notify the Firefighter Assist and Search Team (FAST) deployed in the area of intervention that the wearer of the PASS device is in trouble and in need of rescue. However, this device is not reliable enough since it triggers frequently false positives which lead to developing a tolerance for sounding alarms among the crew. As a consequence, they do not seem to be concerned about it as they should and the alarms are just ignored sometimes. In this paper, we propose a PERsonal LPWAN sYstem (PERLY) prototype for state assessment and localization of Firefighters. The latter's specifications were set by personnel from the fire and emergency response department of the Doubs brigade. The aim was to make the system more reliable compared to the PASS, to add additional important functionalities, and to minimize the false positive alarms.

Mycobacterium tuberculosis complex (MTC) CRISPR locus diversity has long been studied solely investigating the presence/absence of a known set of spacers. Unveiling the genetic mechanisms of its evolution requires a more exhaustive reconstruction in a large amount of representative strains. In this article, we point out and resolve, with a new pipeline, the problem of CRISPR reconstruction based directly on short read sequences in M. tuberculosis. We first show that the process we set up, that we coin as "CRISPRbuilder-TB" (https://github.com/cguyeux/CRISPRbuilder-TB), allows an efficient reconstruction of simulated or real CRISPRs, even when including complex evolutionary steps like the insertions of mobile elements. Compared to more generalist tools, the whole process is much more precise and robust, and requires only minimal manual investigation. Second, we show that more than 1/3 of the currently complete genomes available for this complex in the public databases contain largely erroneous CRISPR loci. Third, we highlight how both the classical experimental in vitro approach and the basic in silico spoligotyping provided by existing analytic tools miss a whole diversity of this locus in MTC, by not capturing duplications, spacer and direct repeats variants, and IS6110 insertion locations. This description is extended in a second article that describes MTC-CRISPR diversity and suggests general rules for its evolution. This work opens perspectives for an in-depth exploration of M. tuberculosis CRISPR loci diversity and of mechanisms involved in its evolution and its functionality, as well as its adaptation to other CRISPR locus-harboring bacterial species.

This paper aims to refine and expand the theoretical and application framework of higher-dimensional digital chaotic system (HDDCS). Topological mixing for HDDCS is strictly proved theoretically at first. Topological mixing implies Devaney's definition of chaos in a compact space, but not vice versa. Therefore, the proof of topological mixing promotes the theoretical research of HDDCS. Then, a general design method for constructing HDDCS via loop-state contraction algorithm is given. The construction of the iterative function uncontrolled by random sequences (hereafter called iterative function) is the starting point of this research. On this basis, this paper put forward a general design method to solve the construction problem of HDDCS, and several examples illustrate the effectiveness and feasibility of this method. The adjacency matrix corresponding to the designed HDDCS is used to construct the chaotic Echo State Network (ESN) for predicting Mackey-Glass time series. Compared with other ESNs, the chaotic ESN has better prediction performance and is able to accurately predict a much longer period of time.

With the advent of the industrial revolution, human beings have developed drastically over the past decades. By 2020, wireless communications would connect more than twenty-five billion devices. Low Power Wide Area (LPWA) technologies are becoming popular as a result of the fast development of the Internet of Things (IoT) market. In this paper, we propose a wearable LoRa-based system for remote safety monitoring of people performing activities in remote areas with no network coverage. The designed system is supposed to detect possible heart problems and/or a “man-down” situation. It then transmits an emergency alert containing information about the state of the concerned individual and its location via LoRa to the surrounding recipients. The proposed system composed of a GPS enabled IoT device, a smart-watch and a smart-phone, has been validated in a remote area in the city of Belfort in France. The obtained results demonstrate the feasibility of such a system.

Many environmental, economic and societal factors are leading fire brigades to be increasingly solicited, and they, therefore, face an ever-increasing number of interventions, most of the time with constant resources. On the other hand, these interventions are directly related to human activity, which itself is predictable: swimming pool drownings occur in summer while road accidents due to ice storms occur in winter. One solution to improve the response of firefighters with constant resources is therefore to predict their workload, i.e., their number of interventions per hour, based on explanatory variables conditioning human activity. The purpose of this article is to show that these interventions can indeed be predicted, in a nonabsurd way, from state-of-the-art tools such as recurrent long short-term memory neural networks (LSTM). From the list of interventions in the Doubs (France), we show that it is possible to build, from scratch, a neural network capable of reasonably predicting the interventions of 2017 from those of 2012-2016. While the results could be improved, they are already promising and would allow the actions of firefighters with a constant resource to be optimized.

Predicting the number and the type of operations by civil protection services is essential, both to optimize on-call firefighters in size and competence, to pre-position material and human resources... To accomplish this task, it is required to possess skills in artificial intelligence, which are not usually found in a medium-sized fire department. However, such a request may be mandated, for example from specialized companies or research laboratories. This mandate requires the transmission of potentially sensitive information relating to interventions which is not intended to be publicly available. The purpose of this article is to show that a machine learning tool can be deployed and provide accurate results, using a learning process based on anonymized data. Learning on real but anonymized data will be performed using extreme gradient boosting, and the performance of each anonymization will be compared on the number and of interventions per day, and their type.

The increase in the world's population and rising standards of living is leading to an ever-increasing number of vehicles on the roads, and with it ever-increasing difficulties in traffic management. This traffic management in transport networks can be clearly optimized by using information and communication technologies referred as Intelligent Transport Systems (ITS). This management problem is usually reformulated as finding the shortest path in a time varying random graph. In this article, an online shortest path computation using stochastic gradient descent is proposed. This routing algorithm for ITS traffic management is based on the online Frank-Wolfe approach. Our improvement enables to find a confidence interval for the shortest path, by using the stochastic gradient algorithm for approximate Bayesian inference. The theory required to understand our approach is provided, as well as the implementation details.

Qualitative but ordered random variables, such as severity of a pathology, are of paramount importance in biostatistics and medicine. Understanding the conditional distribution of such qualitative variables as a function of other explanatory variables can be performed using a specific regression model known as ordinal polytomous regression. Variable selection in the ordinal polytomous regression model is a computationally difficult combinatorial optimisation problem which is however crucial when practitioners need to understand which covariates are physically related to the output and which covariates are not. One easy way to circumvent the computational hardness of variable selection is to introduce a penalised maximum likelihood estimator based on some well chosen non-smooth penalisation function such as, e.g., the l_1-norm. In the case of the Gaussian linear model, the l_1-penalised least-squares estimator, also known as LASSO estimator, has attracted a lot of attention in the last decade, both from the theoretical and algorithmic viewpoints. However, even in the Gaussian linear model, accurate calibration of the relaxation parameter, i.e., the relative weight of the penalisation term in the estimation cost function is still considered a difficult problem that has to be addressed with caution. In the present paper, we apply l_1-penalisation to the ordinal polytomous regression model and compare several hyper-parameter calibration strategies. Our main contributions are: (a) a useful and simple l_1 penalised estimator for ordinal polytomous regression and a thorough description of how to apply Nesterov's accelerated gradient and the online Frank-Wolfe methods to the problem of computing this estimator, (b) a new hyper-parameter calibration method for the proposed model, based on the QUT idea of Giacobino et al. and (c) a code which can be freely used that implements the proposed estimation procedure.

Motivation: The advent of next-generation sequencing has boosted the analysis of bacterial genome evolution. Insertion sequence (IS) elements play a key role in prokaryotic genome organization and evolution, but their repetitions in genomes complicate their detection from short-read data. Results: PanISa is a software pipeline that identifies IS insertions ab initio in bacterial genomes from short-read data. It is a highly sensitive and precise tool based on the detection of read-mapping patterns at the insertion site. PanISa performs better than existing IS detection systems as it is based on a database-free approach. We applied it to a high-risk clone lineage of the pathogenic species Pseudomonas aeruginosa, and report 43 insertions of five different ISs (among which three are new) and a burst of ISPa1635 in a hypermutator isolate. Availability and implementation: PanISa is implemented in Python and released as an open source software (GPL3) at https://github.com/bvalot/panISa. Supplementary information: Supplementary data are available at Bioinformatics online.

BACKGROUND: A huge and continuous increase in the number of completely sequenced chloroplast genomes, available for evolutionary and functional studies in plants, has been observed during the past years. Consequently, it appears possible to build large-scale phylogenetic trees of plant species. However, building such a tree that is well-supported can be a difficult task, even when a subset of close plant species is considered. Usually, the difficulty raises from a few core genes disturbing the phylogenetic information, due for example from problems of homoplasy. Fortunately, a reliable phylogenetic tree can be obtained once these problematic genes are identified and removed from the analysis.Therefore, in this paper we address the problem of finding the largest subset of core genomes which allows to build the best supported tree. RESULTS: As an exhaustive study of all core genes combination is untractable in practice, since the combinatorics of the situation made it computationally infeasible, we investigate three well-known metaheuristics to solve this optimization problem. More precisely, we design and compare distributed approaches using genetic algorithm, particle swarm optimization, and simulated annealing. The latter approach is a new contribution and therefore is described in details, whereas the two former ones have been already studied in previous works. They have been designed de novo in a new platform, and new experiments have been achieved on a larger set of chloroplasts, to compare together these three metaheuristics. CONCLUSIONS: The ways genes affect both tree topology and supports are assessed using statistical tools like Lasso or dummy logistic regression, in an hybrid approach of the genetic algorithm. By doing so, we are able to provide the most supported trees based on the largest subsets of core genes.

Hardware security for an Internet of Things or cyber physical system drives the need for ubiquitous cryptography to different sensing infrastructures in these fields. In particular, generating strong cryptographic keys on such resource-constrained device depends on a lightweight and cryptographically secure random number generator. In this research work, we have introduced a new hardware chaos-based pseudorandom number generator, which is mainly based on the deletion of an Hamilton cycle within the N-cube (or on the vectorial negation), plus one single permutation. We have rigorously proven the chaotic behavior and cryptographically secure property of the whole proposal: the mid-term effects of a slight modification of the seed (proven to be sensitive to the initial conditions) or of the inputted generator cannot be predicted. The proposal has been fully deployed on a FPGA and 65 nm ASIC, it runs completely in parallel while consuming as low resources as possible, and achieving: (a) 11.5 Gb/s for FPGA and 9.4 Gb/s for ASIC random bit throughput, (b) 3.3 μW (LF) to 7.8 mW (UHF) total power consumption with 5% leakage power, measured at 1.32 V, and (c) able to successfully pass the statistical tests of NIST and TestU01 (BigCrush).

In cryptography, Cipher Block Chaining (CBC) mode of operation presents a very popular way of encryption that is used in numerous applications. In our previous research work, we have been proven that, under some conditions, this mode of operation can exhibit a chaotic behavior according to the reputed definition of Devaney. The quantitative study of this chaotic CBC has been deepened later by evaluating both its level of sensibility and of expansivity. In this paper, our objective is now to further develop the evaluation of the CBC dynamics, by obtaining a complete topological study of this mode of operation. Such an evaluation encompasses both the qualitative property of topological mixing and its level of topological entropy, which is indeed a quantitative measure of disorder.

In this paper, we present a toolbox for a specific optimization problem that frequently arises in bioinformatics or genomics. In this specific optimisation problem, the state space is a set of words of specified length over a finite alphabet. To each word is associated a score. The overall objective is to find the words which have the lowest possible score. This type of general optimization problem is encountered in e.g 3D conformation optimisation for protein structure prediction, or largest core genes subset discovery based on best supported phylogenetic tree for a set of species. In order to solve this problem, we propose a toolbox that can be easily launched using MPI and embeds 3 well-known metaheuristics. The toolbox is fully parametrized and well documented. It has been specifically designed to be easy modified and possibly improved by the user depending on the application, and does not require to be a computer scientist. We show that the toolbox performs very well on two difficult practical problems.

Sub-categories of mathematical topology, like the mathematical theory of chaos, offer interesting applications devoted to information security. In this research work, we have introduced a new chaos-based pseudorandom number generator implemented in FPGA, which is mainly based on the deletion of a Hamilton cycle within the $n$-cube (or on the vectorial negation), plus one single permutation. By doing so, we produce a kind of post-treatment on hardware pseudorandom generators, but the obtained generator has usually a better statistical profile than its input, while running at a similar speed. We tested 6 combinations of Boolean functions and strategies that all achieve to pass the most stringent TestU01 battery of tests. This generation can reach a throughput/latency ratio equal to 6.7 Gbps, being thus the second fastest FPGA generator that can pass TestU01.

Motivation: LTR retrotransposons are mobile elements that are able, like retroviruses, to copy and move inside eukaryotic genomes. In the present work, we propose a branching model for studying the propagation of LTR retrotransposons in these genomes. This model allows us to take into account both the positions and the degradation level of LTR retrotransposons copies. In our model, the duplication rate is also allowed to vary with the degradation level. Results: Various functions have been implemented in order to simulate their spread and visualization tools are proposed. Based on these simulation tools, we have developed a first method to evaluate the parameters of this propagation model. We applied this method to the study of the spread of the transposable elements ROO, GYPSY and DM412 on a chromosome of Drosophila melanogaster . Availability and Implementation: Our proposal has been implemented using Python software. Source code is freely available on the web at https://github.com/SergeMOULIN/retrotransposons-spread . Contact: [email protected]. Supplementary information: are available at Bioinformatics online.

The rapid development of medical sensors has increased the interest in Wireless Body Area Network (WBAN) applications where physiological data from the human body and its environment is gathered, monitored, and analyzed to take the proper measures. In WBANs, it is essential to design MAC protocols that ensure adequate Quality of Service (QoS) such as low delay and high scalability. This paper investigates Medium Access Control (MAC) protocols used in WBAN, and compares their performance in a high traffic environment. Such scenario can be induced in case of emergency for example, where physiological data collected from all sensors on human body should be sent simultaneously to take appropriate action. This study can also be extended to cover collaborative WBAN systems where information from different bodies is sent simultaneously leading to high traffic. OPNET simulations are performed to compare the delay and scalability performance of the different MAC protocols under the same experimental conditions and to draw conclusions about the best protocol to be used in a high traffic environment.

Data anonymization is gaining much attention these days as it provides the fundamental requirements to safely outsource datasets containing identifying information. While some techniques add noise to protect privacy others use generalization to hide the link between sensitive and non-sensitive information or separate the dataset into clusters to gain more utility. In the latter, often referred to as bucketization, data values are kept intact, only the link is hidden to maximize the utility. In this paper, we showcase the limits of disassociation, a bucketization technique that divides a set-valued dataset into km-anonymous clusters. We demonstrate that a privacy breach might occur if the disassociated dataset is subject to a cover problem. We finally evaluate the privacy breach using the quantitative privacy breach detection algorithm on real disassociated datasets.

Steganography schemes are designed with the objective of minimizing a defined distortion function. In most existing state of the art approaches, this distortion function is based on image feature preservation. Since smooth regions or clean edges define image core, even a small modification in these areas largely modifies image features and is thus easily detectable. On the contrary, textures, noisy or chaotic regions are so difficult to model that the features having been modified inside these areas are similar to the initial ones. These regions are characterized by disturbed level curves. This work presents a new distortion function for steganography that is based on second order derivatives, which are mathematical tools that usually evaluate level curves. Two methods are explained to compute these partial derivatives and have been completely implemented. The first experiments show that these approaches are promising.

Pseudorandom number generation (PRNG) is a key element in hardware security platforms like fieldprogrammable gate array FPGA circuits. In this article, 18 PRNGs belonging in 4 families (xorshift, LFSR, TGFSR, and LCG) are physically implemented in a FPGA and compared in terms of area, throughput, and statistical tests. Two flows of conception are used for Register Transfer Level (RTL) and High-level Synthesis (HLS). Additionally, the relations between linear complexity, seeds, and arithmetic operations on the one hand, and the resources deployed in FPGA on the other hand, are deeply investigated. In order to do that, a SoC based on Zynq EPP with ARM Cortex-A9 MPSoC is developed to accelerate the implementation and the tests of various PRNGs on FPGA hardware. A case study is finally proposed using chaotic iterations as a post processing for FPGA. The latter has improved the statistical profile of a combination of PRNGs that, without it, failed in the so-called TestU01 statistical battery of tests.

BACKGROUND: Non-Negative Matrix factorization has become an essential tool for feature extraction in a wide spectrum of applications. In the present work, our objective is to extend the applicability of the method to the case of missing and/or corrupted data due to outliers. RESULTS: An essential property for missing data imputation and detection of outliers is that the uncorrupted data matrix is low rank, i.e. has only a small number of degrees of freedom. We devise a new version of the Bregman proximal idea which preserves nonnegativity and mix it with the Augmented Lagrangian approach for simultaneous reconstruction of the features of interest and detection of the outliers using a sparsity promoting ℓ 1 penality. CONCLUSIONS: An application to the analysis of gene expression data of patients with bladder cancer is finally proposed.

In condition-based maintenance, real-time observations are crucial for on-line health assessment. When the monitoring system is a wireless sensor network (WSN), data loss becomes highly probable and this affects the quality of the remaining useful life prediction. In this paper, we present a fully distributed algorithm that ensures fault tolerance and recovers data loss in WSNs. We first theoretically analyze the algorithm and give correctness proofs, then provide simulation results and show that the algorithm is (i) able to ensure data recovery with a low failure rate and (ii) preserves the overall energy for dense networks.

The Internet of Things (IoT) has gained worldwide attention in recent years. It transforms the everyday objects that surround us into proactive actors of the Internet, generating and consuming information. An important issue related to the appearance of such a large-scale self-coordinating IoT is the reliability and the collaboration between the objects in the presence of environmental hazards. High failure rates lead to significant loss of data. Therefore, data survivability is a main challenge of the IoT. In this article, we have developed a compartmental e-Epidemic SIR (Susceptible-Infectious-Recovered) model to save the data in the network and let it survive after attacks. Furthermore, our model takes into account the dynamic topology of the network where natural death (crashing nodes) and birth are defined and analyzed. Theoretical methods and simulations are employed to solve and simulate the system of equations developed and to analyze the model.

In this article, we present a methodology to perform selection among genes based on their expression in various groups of patients, in order to find new genetic markers for specific pathologies. Our approach is based on clustering the denoised data and computing a LASSO (Least Absolute Shrinkage and Selection Operator) estimator, in order to select the relevant genes. This latter belongs to the class of penalized regression estimators where the penalty is a multiple of the ℓ1-norm of the regression vector. Gene markers of the most severe tumor state are finally provided using the proposed approach.

Variations in mitochondrial genes are usually considered to infer phylogenies. However some of these genes are lesser constraint than other ones, and thus may blur the phylogenetic signals shared by the majority of the mitochondrial DNA sequences. To investigate such effects, in this research work, the molecular phylogeny of the genus Taenia is studied using 14 coding sequences extracted from mitochondrial genomes of 17 species. We constructed 16,384 trees, using a combination of 1 up to 14 genes. We obtained 131 topologies, and we showed that only four particular instances were relevant. Using further statistical investigations, we then extracted a particular topology, which displays more robustness properties.

In computational biology and bioinformatics, the manner to understand evolution processes within various related organisms paid a lot of attention these last decades. However, accurate methodologies are still needed to discover genes content evolution. In a previous work, two novel approaches based on sequence similarities and genes features have been proposed. More precisely, we proposed to use genes names, sequence similarities, or both, insured either from NCBI or from DOGMA annotation tools. Dogma has the advantage to be an up-to-date accurate automatic tool specifically designed for chloroplasts, whereas NCBI possesses high quality human curated genes (together with wrongly annotated ones). The key idea of the former proposal was to take the best from these two tools. However, the first proposal was limited by name variations and spelling errors on the NCBI side, leading to core trees of low quality. In this paper, these flaws are fixed by improving the comparison of NCBI and DOGMA results, and by relaxing constraints on gene names while adding a stage of post-validation on gene sequences. The two stages of similarity measures, on names and sequences, are thus proposed for sequence clustering. This improves results that can be obtained using either NCBI or DOGMA alone. Results obtained with this “quality control test” are further investigated and compared with previously released ones, on both computational and biological aspects, considering a set of 99 chloroplastic genomes.

Wireless sensor networks are often deployed in public or otherwise untrusted and even hostile environments, which prompts a number of security issues. Although security is a necessity in other types of networks, it is much more so in sensor networks due to the resource-constraint, susceptibility to physical capture, and wireless nature. In this work we emphasize two security issues: (1) secure communication infrastructure and (2) secure nodes scheduling algorithm. Due to resource constraints, specific strategies are often necessary to preserve the network's lifetime and its quality of service. For instance, to reduce communication costs nodes can go to sleep mode periodically (nodes scheduling). These strategies must be proven as secure, but protocols used to guarantee this security must be compatible with the resource preservation requirement. To achieve this goal, secure communications in such networks will be defined, together with the notions of secure scheduling. Finally, some of these security properties will be evaluated in concrete case studies.

This research work presents a new class of non-blind information hiding algorithms that are stego-secure and robust. They are based on some finite domains iterations having the Devaney's topological chaos property. Owing to a complete formalization of the approach, we prove security against watermark-only attacks of a large class of steganographic algorithms. Finally a complete study of robustness is given in frequency DWT and DCT domains.

As protein folding is a NP-complete problem, artificial intelligence tools like neural networks and genetic algorithms are used to attempt to predict the 3D shape of an amino acids sequence. Underlying these attempts, it is supposed that this folding process is predictable. However, to the best of our knowledge, this important assumption has been neither proven, nor studied. In this paper the topological dynamic of protein folding is evaluated. It is mathematically established that protein folding in 2D hydrophobic-hydrophilic (HP) square lattice model is chaotic as defined by Devaney. Consequences for both structure prediction and biology are then outlined.

This paper introduces a new notion of chaotic algorithms. These algorithms are iterative and are based on so-called chaotic iterations. Contrary to all existing studies on chaotic iterations, we are not interested in stable states of such iterations but in their possible unpredictable behaviors. By establishing a link between chaotic iterations and the notion of Devaney's topological chaos, we give conditions ensuring that these kind of algorithms produce topological chaos. This leads to algorithms that are highly unpredictable. After presenting the theoretical foundations of our approach, we are interested in its practical aspects. We show how the theoretical algorithms give rise to computer programs that produce true topological chaos, then we propose applications in the area of information security.