SUMMARY: Metagenomic workflows involve complex multi-step analyses, from quality control and assembly to binning, annotation, and strain-level profiling. Few existing metagenomic pipelines achieve the combination of flexibility, reproducibility, and hybrid assembly support within a unified workflow. We present StrainMake, a Snakemake-based workflow for de novo metagenomic analysis from short, long, or hybrid sequencing data. StrainMake integrates widely used tools across all major steps-quality control, assembly, binning, dereplication, taxonomic and functional annotation-while also providing non-redundant gene catalogues, community-scale metabolic models, and strain-level microdiversity metrics. The modular design enables the use of alternative tools, scalable execution on HPC systems, and full reproducibility through Snakemake and Conda. RESULTS: Applied to the CAMI II strain-madness dataset, StrainMake produced high-quality assemblies and metagenome-assembled genomes (MAGs), while enabling strain-resolved comparisons across samples. Hybrid assemblies improved contiguity, whereas short-read assemblies offered faster runtimes, illustrating the workflow's benchmarking capacity. AVAILABILITY AND IMPLEMENTATION: StrainMake is open source and available at https://github.com/UMMISCO/strainmake, together with comprehensive documentation. Generated data are deposited in Zenodo (doi: 10.5281/zenodo.16950162).

International audience

The large-scale digitization of herbarium specimens—over 10 million in France via the Recolnat portal led by MNHN, and hundreds of millions worldwide—offers unprecedented opportunities for advancing plant biodiversity research. However, extracting reliable morphological traits from these images remains challenging due to the presence of non-vegetal elements (e.g., labels, rulers, envelopes) and visual background noise that can mislead model predictions. In this work, we introduce a comparative study of three deep learning models—YOLOv8, ViT, and ResNet101—applied to botanical trait classification from digitized herbarium images. We evaluate these models on both raw images and segmented versions where non-plant elements are removed. Although models trained on raw images can sometimes yield high accuracy, interpretability analysis (e.g., Grad-CAM) reveal that their predictions often rely on irrelevant background features. In contrast, segmentation consistently drives the models to focus on the plant itself, leading not only to improved performance for several morphological traits, but also laying the groundwork for explainable and scientifically meaningful AI. We argue that segmentation is not merely a preprocessing step but a prerequisite for trustworthy trait recognition and a necessary condition for enabling AI-driven discovery in botanical sciences. This work establishes a methodological foundation for interpretable plant-focused deep learning applied to biodiversity research.

ABSTRACT We introduce the shape‐image multimodal network (SIM‐Net), a novel 2D image classification architecture that integrates 3D point cloud representations inferred directly from RGB images. Our key contribution lies in a pixel‐to‐point transformation that converts 2D object masks into 3D point clouds, enabling the fusion of texture‐based and geometric features for enhanced classification performance. SIM‐Net is particularly well‐suited for the classification of digitised herbarium specimens—a task made challenging by heterogeneous backgrounds, nonplant elements, and occlusions that compromise conventional image‐based models. To address these issues, SIM‐Net employs a segmentation‐based preprocessing step to extract object masks prior to 3D point cloud generation. The architecture comprises a CNN encoder for 2D image features and a PointNet‐based encoder for geometric features, which are fused into a unified latent space. Experimental evaluations on herbarium datasets demonstrate that SIM‐Net consistently outperforms ResNet101, achieving gains of up to 9.9% in accuracy and 12.3% in F‐score. It also surpasses several transformer‐based state‐of‐the‐art architectures, highlighting the benefits of incorporating 3D structural reasoning into 2D image classification tasks.

At a time when the impacts of climate change and increasing urbanization are making risk management more complex, there is an urgent need for tools to better support risk managers. One approach increasingly used in crisis management is preventive mass evacuation. However, to implement and evaluate the effectiveness of such strategy can be complex, especially in large urban areas. Modeling approaches, and in particular agent-based models, are used to support implementation and to explore a large range of evacuation strategies, which is impossible through drills. One major limitation with simulation of traffic based on individual mobility models is their capacity to reproduce a context of mixed traffic. In this paper, we propose an agent-based model with the capacity to overcome this limitation. We simulated and compared different spatio-temporal evacuation strategies in the flood-prone landlocked area of the Phúc Xá district in Hanoi. We demonstrate that the interaction between distribution of transport modalities and evacuation strategies greatly impact evacuation outcomes. More precisely, we identified staged strategies based on the proximity to exit points that make it possible to reduce time spent on road and overall evacuation time. In addition, we simulated improved evacuation outcomes through selected modification of the road network.

Integer models are of particular interest for applications where predictive models are supposed not only to be accurate but also interpretable to human experts. We introduce a novel penalty term called Facets whose primary goal is to favour integer weights. Our theoretical results illustrate the behaviour of the proposed penalty term: for small enough weights, the Facets matches the L1 penalty norm, and as the weights grow, it approaches the L2 regulariser. We provide the proximal operator associated with the proposed penalty term, so that the regularised empirical risk minimiser can be computed efficiently. We also introduce the Strongly Convex Facets, and discuss its theoretical properties. Our numerical results show that while achieving the state-of-the-art accuracy, optimisation of a loss function penalised by the proposed Facets penalty term leads to a model with a significant number of integer weights.

An important question in microbiology is whether treatment causes changes in gut flora, and whether it also affects metabolism. The reconstruction of causal relations purely from non-temporal observational data is challenging. We address the problem of causal inference in a bivariate case, where the joint distribution of two variables is observed. We consider, in particular, data on discrete domains. The state-of-the-art causal inference methods for continuous data suffer from high computational complexity. Some modern approaches are not suitable for categorical data, and others need to estimate and fix multiple hyper-parameters. In this contribution, we introduce a novel method of causal inference which is based on the widely used assumption that if X causes Y, then P(X) and P(Y|X) are independent. We propose to explore a semi-supervised approach where P(Y|X) and P(X) are estimated from labeled and unlabeled data respectively, whereas the marginal probability is estimated potentially from much more (cheap unlabeled) data than the conditional distribution. We validate the proposed method on the standard cause-effect pairs. We illustrate by experiments on several benchmarks of biological network reconstruction that the proposed approach is very competitive in terms of computational time and accuracy compared to the state-of-the-art methods. Finally, we apply the proposed method to an original medical task where we study whether drugs confound human metagenome.

Score learning aims at taking advantage of supervised learning to produce interpretable models which facilitate decision making. Scoring systems are simple classification models that let users quickly perform stratification. Ideally, a scoring system is based on simple arithmetic operations, is sparse, and can be easily explained by human experts. In this contribution, we introduce an original methodology to simultaneously learn interpretable binning mapped to a class variable, and the weights associated with these bins contributing to the score. We develop and show the theoretical guarantees for the proposed method. We demonstrate by numerical experiments on benchmark data sets that our approach is competitive compared to the state-of-the-art methods. We illustrate by a real medical problem of type 2 diabetes remission prediction that a scoring system learned automatically purely from data is comparable to one manually constructed by clinicians.

To investigate cell spatial organization in complex biological dynamics, an individual-based model that represents cell motion in a deterministic way is proposed and then experimented on avascular tumor growth case. Cell motion remains a fundamental process in many complex biological dynamics such as morphogenesis or cell aggregation. Mathematical models are often used to represent cellular motility and spatial dynamics. However, the variability and the interactions between individuals, which are crucial to build accurate predictive models, are difficult to represent using differential equations. We propose, in this paper, an individual-based approach that allows describing cell motion at deterministic way with diffusion and convection processes. Such approach is built using Smoothed Particle Hydrodynamics (SPH) method and allows representing spatially each cell and its basic biological primitives. The relevance of this approach is illustrated on avascular tumor growth to study the effects of microenvironment (nutrients) and cell individual motion in tumor development. The results shown that the model can qualitatively predict a number of cellular behaviors that have been observed in experiments.

Discovering complex interactions is an important issue in numerous fields ranging from social sciences to systems biology. Over the past few decades, many network learning methods have exhibited competitive results on various types of data. A commonly reached conclusion is that some learning approaches are more advisable than others depending on the dataset type or the complexity of the underlying network. Another frequently encountered issue relates to the ever increasing number of variables that need to be simultaneously dealt with, especially when only a small number of observations is available. The ScaleNet, a novel reconstruction method which can embed different types of network discovery approaches within a spectral framework for large graphical model was introduced recently. The approach identifies sets of connected variables based on the magnitude and sign of the eigenvector elements of a normalized graph Laplacian matrix, and it learns in parallel multiple relevant sub-graphs of a large network. However, the number of eigenvectors to be used and the size of the sub-graphs are to be fixed by an expensive procedure of cross-validation. In this contribution, we propose heuristics to find both optimal number of eigenvectors and the number of nodes in the sub-networks. We illustrate by the results on standard large-scale data sets and on a real human gut graph reconstruction that the proposed approaches save computational time, i.e. are efficient, and reach the state-of-the-art performance.

Score learning aims at taking advantage of supervised learning to estimate interpretable models which facilitate decision making. Ideally, a scoring system is based on simple arithmetic operations, is sparse, and can be easily explained by human experts. In this contribution, we introduce an original methodology to simultaneously learn interpretable binning mapped to a class variable, and the weights associated with these bins contributing to the score. We show by numerical experiments on benchmark data sets that our approach is competitive compared to the state-of-the-art methods. We illustrate by a real medical problem of type 2 diabetes remission prediction that a scoring system learned automatically is comparable to one manually constructed by clinicians.

Feature selection is used to preserve significant properties of data in a compact space. In particular, feature selection is needed in applications, where information comes from multiple heterogeneous high dimensional sources. Data integration, however, is a challenge in itself. In our contribution, we introduce a feature selection framework based on powerful visualisation capabilities of self-organising maps, where the deep structure can be learned in a supervised or unsupervised manner. For a supervised version of the deep SOM, we propose to carry out inference with a linear SVM. A forward-backward procedure helps to converge to an optimal feature set. We show by experiments on real large-scale biomedical data set that the proposed methods embed data in a new compact meaningful representation, allow to visualise biomedical signatures, and also lead to a reasonable classification accuracy compared to the state-of-the-art methods.

BACKGROUND: The last decades witnessed an explosion of large-scale biological datasets whose analyses require the continuous development of innovative algorithms. Many of these high-dimensional datasets are related to large biological networks with few or no experimentally proven interactions. A striking example lies in the recent gut bacterial studies that provided researchers with a plethora of information sources. Despite a deeper knowledge of microbiome composition, inferring bacterial interactions remains a critical step that encounters significant issues, due in particular to high-dimensional settings, unknown gut bacterial taxa and unavoidable noise in sparse datasets. Such data type make any a priori choice of a learning method particularly difficult and urge the need for the development of new scalable approaches. RESULTS: We propose a consensus method based on spectral decomposition, named Spectral Consensus Strategy, to reconstruct large networks from high-dimensional datasets. This novel unsupervised approach can be applied to a broad range of biological networks and the associated spectral framework provides scalability to diverse reconstruction methods. The results obtained on benchmark datasets demonstrate the interest of our approach for high-dimensional cases. As a suitable example, we considered the human gut microbiome co-presence network. For this application, our method successfully retrieves biologically relevant relationships and gives new insights into the topology of this complex ecosystem. CONCLUSIONS: The Spectral Consensus Strategy improves prediction precision and allows scalability of various reconstruction methods to large networks. The integration of multiple reconstruction algorithms turns our approach into a robust learning method. All together, this strategy increases the confidence of predicted interactions from high-dimensional datasets without demanding computations.

Feature selection is an important step when building a classifier. However, the feature selection tends to be unstable on high-dimension and small-sample size data. This instability reduces the usefulness of selected features for knowledge discovery: if the selected feature subset is not robust, domain experts can have little trust that they are relevant. A growing number of studies deal with feature selection stability. Based on the idea that ensemble methods are commonly used to improve classifiers accuracy and stability, some works focused on the stability of ensemble feature selection methods. So far, they obtained mixed results, and as far as we know no study extensively studied how the choice of the aggregation method influences the stability of ensemble feature selection. This is what we study in this preliminary work. We first present some aggregation methods, then we study the stability of ensemble feature selection based on them, on both artificial and real data, as well as the resulting classification performance.

Learning discrete linear functions is a notoriously difficult challenge. In this paper, the learning task is cast as combinatorial optimization problem: given a set of positive and negative feature vectors in the Euclidean space, the goal is to find a discrete linear function that minimizes the cumulative hinge loss of this training set. Since this problem is NP-hard, we propose two simple rounding algorithms that discretize the fractional solution of the problem. Generalization bounds are derived for two important classes of binary-weighted linear functions, by establishing the Rademacher complexity of these classes and proving approximation bounds for rounding methods. These methods are compared on both synthetic and real-world data.

Multi-level agent-based modeling (ML-ABM) requires representing agents at different levels of representation in the same model w.r.t. to time, space and behavior. This paper describes a generic and operational proposal for ML-ABM. First, a generic meta-model for ML-ABM and an associated “morphogenesis” operation are introduced. The generic meta-model allows a modeler to describe multiple levels of representation in the same model, while the “morphogenesis” operation supports agent change of representation level dynamically during the course of the simulation. Second, in order to demonstrate how to operationalize the proposal, we present an implementation of the generic meta-model and the “morphogenesis” operation in the GAMA ABM platform. Finally, we illustrate how our proposal, implemented in the GAMA platform, allows modeler in practice to develop a multi-level agent-based model. To do so, we rely on the famous “Boids” model of Craig Reynolds and show how the modeler can easily introduce a new level of representation of an entity to transform the model into a two-levels agent-based model without having to modify the existing model.

Gene selection is a crucial step when building a classifier from microarray or metagenomic data. As the number of observations is small, the gene selection tends to be unstable. It is common that two gene subsets, obtained from different datasets but dealing with the same classification problem, do not overlap significantly. Although it is a crucial problem, few works have been done on the selection stability. In this paper, we first present some stability quantification methods, then we study the variations of those measures with various parameters (dimensionality, sample size, feature distribution, selection threshold) on both artificial and real data, as well as the resulting classification performance. Feature selection was performed with t-test and classification with linear discriminant analysis. We point out a strong empiric correlation between the dimensionality/sample size ratio and selection instability.

MOTIVATION: The noisy nature of transcriptomic data hinders the biological relevance of conventional network centrality measures, often used to select gene candidates in co-expression networks. Therefore, new tools and methods are required to improve the prediction of mechanistically important transcriptional targets. RESULTS: We propose an original network centrality measure, called annotation transcriptional centrality (ATC) computed by integrating gene expression profiles from microarray experiments with biological knowledge extracted from public genomic databases. ATC computation algorithm delimits representative functional domains in the co-expression network and then relies on this information to find key nodes that modulate propagation of functional influences within the network. We demonstrate ATC ability to predict important genes in several experimental models and provide improved biological relevance over conventional topological network centrality measures. AVAILABILITY: ATC computational routine is implemented in a publicly available tool named FunNet (www.funnet.info).

UNLABELLED: We describe here an exploratory tool, called FunNet, which implements an original systems biology approach, aiming to improve the biological relevance of the modular interaction patterns identified in transcriptional co-expression networks. A suitable analytical model, involving two abstraction layers, has been devised to relate expression profiles to the knowledge on transcripts' biological roles, extracted from genomic databases, into a comprehensive exploratory framework. This approach has been implemented into a user-friendly web tool to promote its open use by the community. AVAILABILITY: http://www.funnet.info

MOTIVATION: Microarray experiments that allow simultaneous expression profiling of thousands of genes in various conditions (tissues, cells or time) generate data whose analysis raises difficult problems. In particular, there is a vast disproportion between the number of attributes (tens of thousands) and the number of examples (several tens). Dimension reduction is therefore a key step before applying classification approaches. Many methods have been proposed to this purpose, but only a few of them considered a direct quantification of transcriptional interactions. We describe and experimentally validate a new dimension reduction and feature construction method, which assesses interactions between expression profiles to improve microarray-based classification accuracy. RESULTS: Our approach relies on a mutual information measure that exposes some elementary constituents of the information contained in a pair of gene expression profiles. We show that their analysis implies a term that represents the information of the interaction between the two genes. The principle of our method, called FeatKNN, is to exploit the information provided by highly synergic gene pairs to improve classification accuracy. First, a heuristic search selects the most informative gene pairs. Then, for each selected pair, a new feature, representing the classification margin of a KNN classifier in the gene pairs space, is constructed. We show experimentally that the interactional information has a degree of significance comparable to that of the gene expression profiles considered separately. Our method has been tested with different classifiers and yielded significant improvements in accuracy on several public microarray databases. Moreover, a synthetic assessment of the biological significance of the concept of synergic gene pairs suggested its ability to uncover relevant mechanisms underlying interactions among various cellular processes.

This paper deals with the possible benefits of perceptual learning in artificial intelligence. On the one hand, perceptual learning is more and more studied in neurobiology and is now considered as an essential part of any living system. In fact, perceptual learning and cognitive learning are both necessary for learning and often depend on each other. On the other hand, many works in machine learning are concerned with "abstraction" in order to reduce the amount of complexity related to some learning tasks. In the abstraction framework, perceptual learning can be seen as a specific process that learns how to transform the data before the traditional learning task itself takes place. In this paper, we argue that biologically inspired perceptual learning mechanisms could be used to build efficient low-level abstraction operators that deal with real-world data. To illustrate this, we present an application where perceptual-learning-inspired metaoperators are used to perform an abstraction on an autonomous robot visual perception. The goal of this work is to enable the robot to learn how to identify objects it encounters in its environment.