Flux sampling and graph neural networks for improved gene essentiality prediction in mammalian genome-scale metabolic models
Kieren Sharma, Lucia Marucci, Zahraa S. Abdallah · 2026
Read Flux sampling and graph neural networks for improved gene essentiality prediction in mammalian genome-scale metabolic models by Kieren Sharma, Lucia Marucci, Zahraa S. Abdallah (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13369363.txt.
Description
Sharma, Marucci, and Abdallah predict which genes a cell cannot live without, using flux sampling and graph neural networks, in a 2026 study. The standard tool, flux balance analysis, assumes the cell pursues a fixed goal, which can bake in the researcher's guess. Flux sampling instead explores all steady flows a metabolic network allows, with no goal imposed. Their model, FluxGAT, learns from those flows laid over the network's graph.
Tested on a hamster cell model and a mouse model with known labels, FluxGAT found more truly essential genes than flux balance analysis. Precision and specificity stayed high. The method shone where the older tool predicted almost no essentials. The authors offer flux-informed networks as a more general route to gene essentiality maps.
Catalog record
| CREATOR | Kieren Sharma, Lucia Marucci, Zahraa S. Abdallah |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |