Fig. 6: Benchmarking performance of yeast-display-trained algorithms. | Nature Communications

Fig. 6: Benchmarking performance of yeast-display-trained algorithms.

From: Repertoire-scale determination of class II MHC peptide binding via yeast display improves antigen prediction

Fig. 6

Receiver operating characteristic (ROC) curves for prediction with existing prediction algorithms, or algorithms trained on our 9mer yeast-display library (YD-trained) or eluted ligand mono-allelic mass spectrometry (MS-trained) data, on either a outlier-removed eluted ligand MS data for HLA-DR401 and -DR402, with expression-matched decoy peptides, or b yeast-display 13mer HLA-DR401 library data, with naïve library decoys. For each dataset, the area under the ROC curve (AUC) and positive predictive value (PPV) of each prediction are shown. Asterisks indicate algorithms that contain the evaluation set in their training data. Source data are provided as a Source Data file.

Back to article page