Skip to main content
. 2021 Jun 1;11(6):496. doi: 10.3390/jpm11060496

Figure 2.

Figure 2

The workflow of the six different integrated human cancer omics datasets into GEM using four integration approaches. We generated 24 types of dataset–algorithm CGEMs. Afterwards, we evaluated the CGEMs by three indices (AFR, EOR, and cancer gene set). Using FBA, we calculated metabolic flux states for all CGEMs and then utilized different statistical and machine learning approaches to find the main metabolic features of CGEMs.