Skip to main content

Table 3 Performance comparison between MSpectraAI and the classic approach utilizing MaxQuant + DNN mode on diverse datasets

From: MSpectraAI: a powerful platform for deciphering proteome profiling of multi-tumor mass spectrometry data by using deep neural networks

  Method PXD007232 [25] PXD008012 [26] PXD007705 [27] PXD005698 [28] PXD002213 [29] PXD009602 [30]
Accuracy Classic 0.70 0.52 0.969 0.625 0.794
MSpectraAI 0.90 0.74 1.00 1.00 1.00 1.00
Sensitivity Classic 0.75 0.551 1.00 0.615 0.765
MSpectraAI 1.00 0.727 1.00 1.00 1.00 1.00
Precision Classic 0.60 0.593 0.938 0.667 0.813
MSpectraAI 0.833 0.696 1.00 1.00 1.00 1.00
F1 Classic 0.667 0.571 0.968 0.64 0.788
MSpectraAI 0.909 0.711 1.00 1.00 1.00 1.00
  1. “Classic” in the Method means the classic approach- “MaxQuant + DNN” mode; –, means no value