Fig. 5From: MSpectraAI: a powerful platform for deciphering proteome profiling of multi-tumor mass spectrometry data by using deep neural networksa F1-score distribution of MS1 mass spectra (marked with dots) and MS2 mass spectra (marked with triangles) across every sample. b ROC curves for each tumor type calculated using MS1 data and c MS2 data. d Accuracy of the data prediction of every tu-mor type. e Heatmap of the multiclassification prediction probability of every sample (Class 1: Normal samples; Classes 2 and 3: two different treatment conditions of nonsmall cell lung cancer samples)Back to article page