Skip to main content
. 2024 Mar 28;16(1):2333436. doi: 10.1080/19420862.2024.2333436

Table 3.

Metrics for models predicting Asn degradation in validation set. The letters A, B and C refer to the different sets of predictors. ‘A’ represents the SASA-group (including ASA and ASA(n + 1)), ‘B’ denotes the Static-group (including ASA, ASA(n + 1), Amide_pka, and Area_amide), and ‘C’ refers to the Dynamic-group (including ASA, SC_rmsd, Ca_rmsd, and Amide_pKa). The sequence-based approach classifies NG and NS as the only two motifs susceptible to degradation.10

  Homology
Model
Low MD
50 Confs.
Low MD
400 Confs.
ST
pH6
ST
pH8.5
MD Sim
200ns
Seqence-
based
ST pH8.5 Thres 0.3
Predictor A B A B C A B C C C C NG, NS C
TPR (Sensitivity, %) 40.7 47.3 45.3 58.0 47.3 49.3 53.3 62.0 50.7 58.0 50.0 52.0 80.0
TNR (Specificity, %) 82.7 71.6 83.1 83.1 76.9 84.4 82.7 80.9 82.2 83.6 85.8 90.2 72.4
Error (%) 34.1 38.1 32.0 26.9 34.9 29.6 29.1 26.7 30.4 26.7 28.5 25.1 24.5
Precision (%) 64.3 52.6 64.2 69.6 57.7 71.6 67.2 68.4 65.5 70.2 70.1 78.0 66.3
F1-score (%) 48.8 49.8 53.1 63.3 52.0 56.5 59.5 65.0 57.1 63.5 58.4 62.4 72.1
Balanced accuracy (%) 61.7 59.4 64.2 70.6 62.1 66.8 68.0 71.4 66.4 70.8 67.9 71.1 76.2
Total accuracy (%) 65.9 61.9 68.0 73.1 65.1 70.4 70.9 73.3 69.6 73.3 71.5 74.9 75.5
AUC in (0,1) 0.72 0.63 0.74 0.75 0.70 0.74 0.77 0.80 0.75 0.84 0.77 0.71 0.84