Mots-Clés
lncRNA
plants
Artificial inteligence
LLM
embedding
Description
A two-year Postdoctoral position funded by the French-Korean (ANR-NRF) bilateral DHARP project is available in the “Functional conservation of plant lncRNAs” (FunRNA) team, led by Thomas Blein and Jérémie Bazin at the Institute of Plant Sciences Paris-Saclay (IPS2), France.
Long non-coding RNAs (lncRNAs) have long been regarded as transcriptional noise due to their low protein coding potential, their poor conservation and their low level of expression. However, an increasing number of studies have shown that lncRNAs can regulate gene expression at different levels and play an important role in a variety of biological processes. Our lab investigates the molecular features of lncRNAs, examining their expression variability across different conditions and species, as well as their sequence conservation.
This project focuses on solving a major challenge in plant science: the functional annotation of lncRNAs, focusing on the ones involved in rice drought and heat tolerance. Indeed, despite the growing evidence of their functional significance, understanding the roles of plant lncRNAs is challenging due to their low sequence conservation and lack of canonical motifs, which complicates functional annotation. While thousands of lncRNAs are differentially expressed under stress conditions, distinguishing functional molecules from transcriptional noise remains a major hurdle. Our central hypothesis is that functional lncRNAs can be identified through the integration of large-scale omics datasets with AI-driven analysis of RNA sequence and structure. The project aimed at applying AI and deep-learning foundation models to systematically characterize these molecules. This involves generating high-dimensional numerical vector embeddings that capture rich structural and contextual features of the lncRNAs, which are then used to cluster them into groups based on predicted functional similarity. The different groups will then be annotated using omics datasets, such as transcriptomics, epigenomics, subcellular localization, and RNA-binding protein profiles, to identified the most promising candidate for further characterization.
We are looking for talented, highly motivated and creative researchers with a PhD in Computational Biology, Bioinformatics, or a related field. Candidates should have experience with AI/Deep Learning methodologies and their application to biological sequence data. Candidates should have a solid background handling and analyzing large sequencing datasets, including functional genomics data processing including epigenomic data. The candidate will work in an interdisciplinary and multinational team. This dynamic environment provides a supportive and stimulating atmosphere conducive to research excellence. Furthermore, the team favor teamwork and professional development.