Fig. 2
From: Molecular typing of gliomas on the basis of integrin family genes and a functional study of ITGA7
illustrates the process of constructing a glioma model. (A) Presents the performance of Non-negative Matrix Factorization (NMF) under different factorization ranks. It includes six subplots, each displaying the variation of six indicators: cophenetic, dispersion, evar, residuals (RSS), silhouette, and sparseness, with respect to the factorization rank. Each indicator showcases the performance of different decomposition methods, including Basis, Best fit, Coefficients, and Consensus. (B) Presents the consensus matrix when the fractalization is set to 2. (C) Depicts the survival differences observed in the typing. Independent external datasets as validation sets, including GEO16011 (D), CGGA301 (E), TCGA (see F), and GEO18494 and GEO4412 (G). (H) Shows the proportion of samples between the classification and the WHO grade. (I) features a differential volcano diagram of the classification. Finally, (J) displays the KEGG results for the typing.
