Figure 4
From: Comprehensive application of AI algorithms with TCR NGS data for glioma diagnosis
(A,B) represent the distribution of importance scores obtained from Lasso with threshold values of 0.1 and 0.001, respectively. (C,D) demonstrate the optimal performance of RFECV, indicating that 5 TCRβ sequences yield the highest AUC. (E,F) reveal the ROC curves of the two groups of core sequences and phenotype. (G,H) depicts the correlation coefficient plots of the two groups of core sequences. (I) shows the seqlogo image representing the core 3 TCRβ sequences obtained from the RFECV feature selection algorithm.
