FAIROmics - PhD fellowship in the design of Information Extraction Tools to characterise molecules produced or degraded by microbes and applications to plant-fermented food ecosystems.

"FAIRification of multiOmics data to link databases and create knowledge graphs for fermented foods" MSCA-DN-JD Doctoral Network.

The FAIROmics initiative, an interdisciplinary research programme, will gather universities, research centres and private companies to enable the FAIRification of omics data and databases interoperability and develop knowledge graphs for data-driven decision-making to rationally design microbial communities for imparting desirable characteristics to plant-based fermented foods in the context of open science and its regulations. The FAIROmics training programme aims to develop doctoral candidates’ skills at the interface between artificial intelligence, life sciences, humanities, and social sciences.     

Plant-based dairy and meat alternatives have grown in popularity in recent years for various reasons, including sustainability and health benefits, as well as lifestyle trends and dietary restrictions. However, plant-based food products can be nutritionally unbalanced, and their flavour profiles may limit their acceptance by consumers. Microorganisms have been used in making food products for millennia. However, the diversity of microbial communities driving plant-based fermentations, as well as their key genetic and phenotypic traits and potential synergies among community members, remain poorly characterised. Many data exist, but they are spread into different literature (scientific and grey) or, in the best case, in different databases. However, they are not always reusable because they are difficult to find and access and because databases are not systematically interoperable.

Please note that this PhD position will lead to the award of a double diploma after the completion of a stay in each of these organisations: The University of Paris-Saclay (UPSaclay), France and the University of Szeged (USZ), Hungary.

Objectives:

We are looking for one Doctoral Candidate (DC) to join our project at multiple sites in the EU with a master’s degree in a relevant discipline (Master’s degree in engineering, physics, systems biology, applied mathematics, biotechnology) interested in Modelling, analysis and control of biological systems in the context of microbial fermentations.

The PhD project aims to develop information extraction (IE) methods to automatically produce a knowledge graph about microbe biology involved in plant-based food transformation or preservation. The knowledge graph will formalise the molecules produced and degraded by microorganisms in the fermentation process.

The IE methods will involve named-entity recognition, entity normalisation with respect to semantic references and relationship extraction. They will be based on the most recent deep learning approaches that train language models using few or no training examples by transfer learning or exploiting existing structured information, i.e. knowledge bases and ontologies for distant or weak learning by including relevant information according to the needs of the FAIROmics dedicated use cases (e.g. NCBI Taxonomy for taxa, FoodEX2 for food, ChEBI for molecules, KEGG for pathways). Existing annotated corpora will serve as a starting point for training (e.g. CHEMDNER, Pathway Curation, Bacteria Biotope).

The project will rely on existing tools and resources on microbe biology developed by MaIAGE partners (e.g. Omnicrobe application*, Ontobiotope ontology*, extraction workflow).

Expected results:

The PhD student will design and evaluate original machine-learning-based methods for extracting information on plant-based fermentation metabolism from text. The models and software will be available to the scientific community in an open-source license. The extracted knowledge will feed a publicly available knowledge graph of microbial properties. The results will be published in the major NLP venue and relevant bioinformatics journals.

Location and planned secondments:

The PhD student will be mainly located at the INRAE site in Jouy-en-Josas for 24 months and at the Szeged University for a 12-month secondment.

Enrolment in Doctoral degree:

1st-degree awarding organisation: University Paris-Saclay, https://www.universite-paris-saclay.fr/
2nd degree awarding organisation: University of Szeged, https://u-szeged.hu/english

Supervisors team

French team:
Two MaIAGE teams will be involved in the PhD supervision: the Bibliome team* and the StatInfOmics team*:

• Robert Bossy (Bibliome): Natural Language Processing and application to microbiology, software engineering.
• Claire Nédellec (Bibliome): Natural Language Processing and application to microbiology, knowledge representation and ontology.
• Hélène Chiapello (StatInfOmics): Microbial bioinformatics, omics data.
• Sandra Dérozier (StatInfOmics): Microbial bioinformatics, software engineering.

Hungarian team :

• Vidács Lázló: Artificial intelligence, natural language processing, software engineering.
• Balázs Nagy: Artificial intelligence, natural language processing, software engineering.

Host institutions description

INRAE is Europe’s top agricultural research institute and the world’s number two centre for the agricultural sciences. Its scientists are working towards solutions for society’s major challenges. RU1404 MaIAGE gathers mathematicians, computer scientists, bioinformaticians and biologists to tackle problems from biology, agronomy and ecology. Our research concerns processes at various levels, ranging from molecular, cellular or multicellular levels to organisms, populations, and entire ecosystems.

The University of Szeged (USZ) is recognised as a top research institution in Hungary, boasting a diverse student body of over 21,000, including more than 4,000 international students from 115 countries. Led by László Vidács, the Applied Artificial Intelligence Research Group is dedicated to advancing cutting-edge AI research. We specialise in diverse AI applications, from natural language understanding to image processing. Our tailored machine learning and deep learning solutions address real-world challenges in many domains, including medical imaging diagnostics, forensic text analysis, and program source code processing.