INFERRING METABOLIC NETWORKS FROM TRANSCRIPTOMIC DATA: A GATEWAY TO UNDERSTAND FATE CONTROL BY METABOLISM IN THE EPIBLAST
Transcriptomics data have allowed unprecedented advances to our understanding of cell fate transitions during human development. A bottleneck in the analyses lies in our ability to infer biological functions from genomic and transcriptomic data. Here, we plan to develop a framework to leverage transcriptomics data from human pluripotent stem cells in the primed, and naive states to infer metabolic networks. In a proof of concept study (manuscript on biorxiv), we highlight the value of weighting genes based on the metabolic network for the identification of key metabolic characteristics of stem cell populations. Here, we plan to integrate metabolic gene regulatory networks (GRN) to link cell fate decisions and metabolism. To achieve this, we will work on three aims. The first aim is to compare metabolic networks computed from transcriptomic or proteomic data. In order to assess the robustness of our metabolic network, we will compare networks computed from human data and cynomolgus macaque data (transcriptomics and proteomics). The second aim is to develop a novel framework that integrates GRNs into the inference of metabolic networks. This will involve analyzing scRNA-seq data to identify key genes, followed by the inference of GRNs using a newly designed algorithm that incorporates prior knowledge of gene regulation. By leveraging these GRNs, we will identify active pathways influencing metabolic processes and predict how gene perturbations could impact metabolism. The third aim is to validate the metabolic networks by perturbing metabolic reactions in human and cynomolgus pluripotent stem cells states. Altogether, our project capitalizes on expertise in stem cell biology, evo-devo in cynomolgus vs human, metabolic network computing and scRNAseq inference of cell fate regulatory networks to advance developmental biology.
3/6 complete
Imported from OpenAIRE Germany | Grant ANR-25-CE13-5256.
Open sourceOverview
Grant agreement ANR-25-CE13-5256 | METAGRiN
Open sourceTranscriptomics data have allowed unprecedented advances to our understanding of cell fate transitions during human development. A bottleneck in the analyses lies in our ability to infer biological functions from genomic and transcriptomic data. Here, we plan to develop a framework to leverage transcriptomics data from human pluripotent stem cells in the primed, and naive states to infer metabolic networks. In a proof of concept study (manuscript on biorxiv), we highlight the value of weighting genes based on the metabolic network for the identification of key metabolic characteristics of stem cell populations. Here, we plan to integrate metabolic gene regulatory networks (GRN) to link cell fate decisions and metabolism. To achieve this, we will work on three aims. The first aim is to compare metabolic networks computed from transcriptomic or proteomic data. In order to assess the robustness of our metabolic network, we will compare networks computed from human data and cynomolgus macaque data (transcriptomics and proteomics). The second aim is to develop a novel framework that integrates GRNs into the inference of metabolic networks. This will involve analyzing scRNA-seq data to identify key genes, followed by the inference of GRNs using a newly designed algorithm that incorporates prior knowledge of gene regulation. By leveraging these GRNs, we will identify active pathways influencing metabolic processes and predict how gene perturbations could impact metabolism. The third aim is to validate the metabolic networks by perturbing metabolic reactions in human and cynomolgus pluripotent stem cells states. Altogether, our project capitalizes on expertise in stem cell biology, evo-devo in cynomolgus vs human, metabolic network computing and scRNAseq inference of cell fate regulatory networks to advance developmental biology.