The TNT-Sensor-IA project aims in designing of a new generation of multiplexed array microsensors integrating a metamaterials whose porosity can be spatially programmed. Such a direct laser writing approach associates the latest advances in multiphoton stereolithography (SLA) with those in the artificial intelligence. A progressive sensor specialization processing based on deep learning methods will be performed in order to detect TNT (2,4,6-trinitrotoluene) traces in complex gas mixtures. Note that the majority of the sensors use machine learning only through a unidirectional procedure that provides a simple feed back analysis of the sensor performance. The originality of our project is to propose an iterative algorithmic strategy to reprogram the sensor after each learning cycle. This advantage stems from the computational flexibility proposed by SLA to reprogram on demand the porosity of this metamaterial over a wide space of possible configurations. Hence, beyond the strategic issue to target TNT, the principal constituent of 85 % of unexploded land mines worldwide, the great ability of our multiplexed sensors for learning make them very promising candidates for other sensing applications.

Existing electrochemical sensors for environmental monitoring suffer from limitations in terms of sensitivity and selectivity. Better sensors are needed to respond to the rising societal demand for continuous information on environmental safety. The HYPERION project will investigate the development of functionalised membranes hierarchically porous to form a new class of electrochemical sensors based on ion transfer voltammetry. The sensors will consist of a macroporous polymeric membrane modified electrochemically with a mesoporous silica film. This project will focus on (i) the preparation of hierarchically porous membranes using sol-gel chemistry and micro-fabrication methods, (ii) the understanding of the formation mechanism and (iii) the exploitation of the selectivity properties of the pore dimension and of the surface chemistry for analytical applications. The research developed in this project will combine (i) microscopic features to improve the mass transport and (ii) mesopores with chemical functions to boost both sensibility and selectivity. This project will establish the fundamental knowledge for the development of a novel class of electrochemical sensors.

The use of renewable resources is essential for a sustainable society. Developing clean catalytic processes to produce value-added chemicals from renewable materials such as wood or plants has become a major goal for chemists. The bio-feedstocks issued from lignocellulose after either enzymatic fermentation or acidic deconstruction consist mainly of water-soluble molecules containing many oxygenated groups ((di)acids, alcohols, ethers), which must be subsequently transformed to find applications as monomers, solvents, etc. For this purpose, the design of new water-stable catalysts able to withstand rather harsh reaction conditions in term of pH, temperature and pressure is required. The NHYSCAB project aims at the synthesis of hydrothermally stable promoted metallic catalysts, based on a supported noble metal (e.g. Pd, Ru) modified with an oxophilic promoter (Re or Mo). They will be used for catalytic hydrogenation/hydrogenolysis of biosourced molecules in aqueous phase, at temperatures in the range 100 to 200°C and hydrogen pressure in the range 50 to 150 bar, at acidic to neutral pH. This project is based on the collaboration between two complementary public partners, IRCELYON (coordinator) and ICGM Montpellier, which are internationally recognized for their expertise in the fields of biomass catalytic upgrading and of non-hydrolytic sol-gel (NHSG) synthesis of mixed oxides, respectively. The first part of the project consists in the design of advanced mesoporous catalyst supports using the NHSG process which offers powerful synthetic routes to hydrothermally stable mesoporous oxides (TiO2, ZrO2) and mixed oxides incorporating the promoter species (Re-Ti, Mo-Ti), which after an appropriate thermal treatment are dispersed at the surface of the oxide. The hydrothermal stability of these supports will be assessed under the reaction conditions. Noble metal will be deposited on the stable supports with well-defined compositions and structures in order to prepare efficient promoted metallic catalysts. The second part concerns the evaluation of the synthesized catalysts in the reference reaction of aqueous-phase hydrogenation of biosourced acids (succinic acid and levulinic acid) to the corresponding diols (1,4-butane- and 1,4-pentane- diols). The catalysts will also be evaluated in the challenging hydrogenolysis of tetrahydrofurfuryl alcohol from the furfural platform into the corresponding 1,2- or 1,5- pentanediols. The expected products can find many applications, including as monomers. The design of well-defined, thoroughly characterized solids is essential to optimize the selective synthesis of targeted chemicals. Therefore, extensive characterization of the solids (supports and catalysts) will be performed at different stages (oxides, mixed oxides, supported metallic catalysts, before and after reaction) and their stability will be investigated under the reaction conditions. These will allow us to determine the texture/structure/composition of the solids, to validate their stability and to correlate the characteristics of the catalysts and their performance. To reach the highest activity or selectivity, the study will focus not only on the catalyst design but also on the optimization of reaction conditions. After screening of catalyst compositions using a batch reactor, the catalytic reaction will be conducted in a continuous trickle-bed reactor to further study the stability of the selected catalytic systems.

Anaerobic digestion (AD) is a microbiological process of degradation of the organic matter which produces biogas rich in methane that can be converted into valuable electrical and thermal energy. It is commonly used to manage different types of organic waste at industrial scale using anaerobic digesters. However, this bioprocess is not fully mastered and still has an important potential for improvement. One of the major limitations of AD is the important susceptibility of the microbial communities to changes in operational conditions of the digesters. It can lead to unstable methane formation. Controlling AD microbial community stability, though, is not a trivial task. Knowledge on the determinants of anaerobic microbial process stability (i.e. the conditions and the succession of microbial events that allow maintaining a balance after a disruption or, on the contrary, that generate a domino effect leading to total failure) over time is still missing. Emerging omics high-throughput approaches can now lead to unprecedented data to portray AD microbiome. Metagenomics, metatranscriptomics, metaproteomics and metabolomics enable to describe a community at different levels (genes, gene expression, and metabolites production). Appropriate and efficient analytical methods are required to analyse these big and complex data and unravel the intricate networks of functional processes of AD. Novel computational and statistical methods are progressively becoming available to fully harvest and integrate these complex datasets. In this context, the aim of STABILICS is to conduct the first sets of high-throughput multi-omics longitudinal experiments, with an unprecedented sampling depth, in anaerobic digesters under constant environmental parameters or subject to different model perturbations created by the addition of NaCl. Experiments in lab-scale semi-continuous reactors will be set-up and monitored in the long run (more than one year). Two levels of analysis will be applied. 1) A high frequency monitoring of different descriptors of microbiota activity, where non-targeted metabolomics and isotopic analyses will characterise the degradation pathways and metabarcoding of RNA and DNA will target both active and present microorganisms. 2) An in-depth monitoring of microbiota functioning with both metagenomics and metatranscriptomics on selected samples and conditions. These unprecedented sets of data will be thoroughly analysed and integrated using cutting-edge statistical methods. For example, multivariate dimension reduction methods will be used for data mining, omics integration and feature selection; specific analytical framework for longitudinal data will be developed. The objectives of this interdisciplinary project will be 1) to evaluate at different omics levels the dynamics of AD microbiome in long term and replicated time course experiments, 2) to describe the succession of events that, under stress, leads to microbiota equilibrium unbalance and digester disruption or on the contrary microbiota equilibrium preservation and maintenance of stability, 3) to propose an original analytical framework of multi-omics longitudinal studies accounting for temporality, and 4) to deliver generic knowledge to understand the determinants of perturbations.

Ribonucleoprotein particles (RNPs) are made of RNA associated with proteins and play a central role in biological systems (maintenance of cellular homeostasis, establishment of infectious or pathological processes). The identification of the components of these RNPs has exploded over the past decade thanks to the use of high-throughput analytical approaches. The fine characterization of the interaction of the components of these RNPs then requires the analysis of a large number of mutants. While several high-throughput methodologies have been developed for the analysis of RNA mutant libraries, progress has been scarcer on the protein side. To fill this gap, we propose "SURF", a highly multidisciplinary project aiming at developing a new chemistry for the efficient capture of target RNAs on the surface of water-in-oil droplets produced, manipulated, and analyzed at rates of several million per hour in microfluidic devices. Gene libraries encoding mutants of the studied protein fused to a fluorescent domain will be expressed in vitro at a rate of one mutant (produced in large numbers of copies) per droplet. Thus, a mutant able to interact with the target RNA sequence will lead to a relocation of the fluorescent protein on the surface of the droplet, making it easily discriminable from a drop containing a mutant unable to recognize its target (fluorescence remaining diffuse in the droplet). Applied and validated with various biological models, this technology will not only allow to finely characterize the formation of RNPs, but also to reprogram their specificity, and even to identify molecules able to modulate this interaction. Finally, this project will also be an opportunity to explore surfactants made of alternative chemistries having a lower impact on the environment.