Fig. 4: Benchmark comparison of composite model ImmuneApp-MA that integrates both mono and multi-allelic immunopeptidomics data. | Nature Communications

Fig. 4: Benchmark comparison of composite model ImmuneApp-MA that integrates both mono and multi-allelic immunopeptidomics data.

From: ImmuneApp for HLA-I epitope prediction and immunopeptidome analysis

Fig. 4

a. AUROC, AUPRC, and PPV values for each allele in the external mono-allelic EL testing dataset for ImmuneApp-MA, ImmuneApp-EL, NetMHCpan-4.1, MHCflurry 2.0, MixMHCpred 2.1&2.2, HLAthena, MHCnuggets-2.4 and TransPHLA. b. Mean AUROC, AUPRC, and PPV values with a 95% confidence interval (CI) stratified by both HLA and epitope length (n = 96) were calculated, followed by a two-tailed Wilcoxon signed-rank test for adjusted P-values. Source data are provided as a Source Data file.

Back to article page