Intégration d'outils analytiques et d'apprentissage automatique pour identifier des marqueurs et prédire la réactivité de la biomasse lignocellulosique en hydrolyse enzymatique.

L'hydrolyse enzymatique (HE) est l'une des voies de conversion de la biomasse lignocellulosique de seconde génération en éthanol ou en molécules plateformes. Aujourd'hui, étant donné le manque de compréhension de la relation entre les propriétés physiques et chimiques de la biomasse et sa réactivité et le manque d'outils de prédiction, il est toujours nécessaire de réaliser des expériences coûteuses dans des conditions représentatives.

Pour envisager une meilleure rentabilité des procédés de conversion de la biomasse lignocellulosique par la voie biochimique, il devient essentiel d'établir le lien entre les propriétés physiques et chimiques de la biomasse lignocellulosique et le rendement en sucre observé.

Amaretto - Schema projet
Organisation du projet ciblé Amaretto


Les objectifs du projet :

Méthodologie analytique

Proposer et évaluer une méthodologie analytique :

  • Sélection d’une série d’échantillons présentant une variabilité suffisante ;
  • Approches multi-techniques d’analyse des hydrolysats : détail moléculaire de mélanges complexes ;
  • Approches multi-techniques d’analyse des marcs PT, aux différentes échelles ;
  • Mise en forme et structuration du jeu de données multidimensionnelles et hétérogènes ;
  • Association d’une donnée de réactivité à chaque individu.
Identification de marqueurs

Identifier des marqueurs de réactivité :

  • Marqueurs d’inhibition de l’HE présents dans les hydrolysats ;
  • Propriétés clés du solide qui régissent la réactivité en HE
Proposer un modèle de prédiction de la réactivité


Le consortium :

Établissements d'enseignement supérieur
Instituts de recherche


HAL : Dernières publications

  • [hal-04644779] Exploring the richness of the French Galaxy Ecosystem

    The French Bioinformatics Community has embraced Galaxy since its inception, with a pivotal moment being the Galaxy Tour de France led by Nate Coroar, Anton Nekrutenko, and James Taylor in 2012. This adoption has led to the establishment of over 10 Galaxy servers across France, catering to diverse local needs and specialized thematic areas such as ecology, biodiversity, NGS, proteomics, and more. Among these servers, stands out as the flagship national instance, launched in 2021 and hosted by the French Institute for Bioinformatics (IFB - ELIXIR-FR). With robust infrastructure boasting 8300 CPU cores, 52 TB of RAM, and GPU cards, offers a comprehensive suite of over 3,000 tools, including interactive options like Jupyter Notebook, AlphaFold, and Helixer. Notably, it has garnered over 6,000 users who have collectively executed over 3.6 million jobs. Moreover, hosts specialized subdomains catering to various community needs, such as ecology, metabarcoding, and COVID-19 research, with ongoing integration of new subdomains. The community's commitment to collaboration and consolidation is evident as several local servers have migrated to in recent years, with others expressing interest in doing the same. The French Galaxy community is deeply engaged in a multitude of projects at national, European, and global levels, including EOSC FAIR EASE, EuroScienceGateway, ATLASea and ABRomics. To foster cohesion and synergy within the community, a Galaxy Working Group led by the French Bioinformatics Institute facilitates regular interactions. This group serves to connect Galaxy users across France, share knowledge, support, and combat misconceptions about Galaxy within the French scientific community. In this poster presentation, we provide an overview of the dynamic French Galaxy ecosystem, highlighting its diverse servers, engaged researchers, ongoing projects, and collaborative efforts. Through this exploration, we aim to showcase the vibrancy and impact of Galaxy within the French bioinformatics landscape. (Bérénice Batut) 11 Jul 2024
  • [hal-04281075] Dryland endolithic Chroococcidiopsis and temperate fresh water Synechocystis have distinct membrane lipid and photosynthesis acclimation strategies upon desiccation and temperature increase

    An effect of climate change is the expansion of drylands in temperate regions, predicted to affect microbial biodiversity. Photosynthetic organisms being at the base of ecosystem’s trophic networks, we compared an endolithic desiccation-tolerant Chroococcidiopsis cyanobacteria isolated from gypsum rocks in the Atacama Desert, with a freshwater desiccation-sensitive Synechocystis. We sought whether some acclimation traits in response to desiccation and temperature variations were shared, to evaluate the potential of temperate species to possibly become resilient to future arid conditions. When temperature varies, Synechocystis tunes the acyl composition of its lipids, via a homeoviscuous acclimation mechanism known to adjust membrane fluidity, whereas no such change occurs in Chroococcidiopsis. Vice versa, a combined study of photosynthesis and pigment content shows that Chroococcidiopsis remodels its photosynthesis components and keeps an optimal photosynthetic capacity at all temperatures, whereas Synechocystis is unable to such adjustment. Upon desiccation on a gypsum surface, Synechocystis is rapidly unable to revive, whereas Chroococcidiopsis is capable to recover after three weeks. Using X-ray diffraction, we found no evidence that Chroococcidiopsis could use water extracted from gypsum crystal in such conditions, as a surrogate of missing water. The sulfolipid sulfoquinovosyldiacylglycerol becomes the prominent membrane lipid in both dehydrated cyanobacteria, highlighting an overlooked function for this lipid. Chroococcidiopsis keeps a minimal level of monogalactosyldiacylglycerol, which may be essential for the recovery process. Results support that two independent adaptation strategies have evolved in these species to cope with temperature and desiccation increase, and suggest some possible scenarios for microbial biodiversity change triggered by climate change. (Damien Douchi) 12 Nov 2023

Date de modification : 08 juillet 2024 | Date de création : 08 novembre 2023 | Rédaction : B-BEST