scFLUX: a web server for metabolic flux and variation prediction using transcriptomics data [article]

Zixuan Zhang, Wennan Chang, Norah Alghamdi, Haiqi Zhu, Mengyuan Fei, Changlin Wan, Alex Lu, Yong Zang, Ying Xu, Wenzhuo Wu, Sha Cao, Yu Zhang (+1 others)
2022 bioRxiv   pre-print
Quantitative assessment of single cell fluxome is critical for understanding the metabolic heterogeneity in diseases. Unfortunately, single cell fluxomics using laboratory approaches is currently infeasible, and none of the current flux estimation tools could achieve single cell resolution. In light of the natural associations between transcriptomic and metabolomic profiles, it remains both a feasible and urgent task to use the available single cell transcriptomics data for prediction of single
more » ... cell fluxome. We present scFLUX here, which provides an online platform for prediction of metabolic fluxome and variations using transcriptomics data, on individual cell or sample level. This is in contrast to other flux estimation methods that are only able to model the fluxes for cells of pre-defined groups. The scFLUX webserver implements our in-house single cell flux estimation model, namely scFEA, which integrates a novel graph neural network architecture with a factor graph derived from the complex human metabolic network. To the best of our knowledge, scFLUX is the first and only web-based tool dedicated to predicting individual sample-/cell- metabolic fluxome and variations of metabolites using transcriptomics data. scFLUX is available at http://scflux.org/. The stand-alone tools for using scFLUX locally are available at https://github.com/changwn/scFEA.
doi:10.1101/2022.06.18.496660 fatcat:wlxlsx3tpjdltdgd7ugh52jooa