Fig. 9: Machine-learning frameworks, prognostic modeling, and conceptual summary of lactate-driven microenvironmental remodeling. | npj Digital Medicine

Fig. 9: Machine-learning frameworks, prognostic modeling, and conceptual summary of lactate-driven microenvironmental remodeling.

From: Machine learning-enabled spatial multi-omics uncovers lactate-driven targets and tumor microenvironmental reprogramming in cancer

Fig. 9: Machine-learning frameworks, prognostic modeling, and conceptual summary of lactate-driven microenvironmental remodeling.The alternative text for this image may have been generated using AI.

a Random forest classifier distinguishing high- versus low-lactate tumors based on cell-type composition. b Mean−Squared Error statistics linking spatial lactate intensities to gene expression module scores. c Elastic-net model coefficients highlighting a minimal set of lactate-associated gene modules. d Support vector machine (SVM) classifier identifying the top discriminative genes for lactate status, including PDHB, HK2, EPAS1, MYC, and SLC2A1. e Artificial neural network (ANN) model trained on deconvolved spatial transcriptomics. f ROC curve for the elastic-net model predicting high-lactate regions. g Decision-tree classifier trained on TCGA lactate-associated gene expression. h Graphical summary model summarizing the study’s findings.

Back to article page