Data Supplement for The Journal of Immunology: Volume 169, Issue 10, Page 5744
Udaka. et al.: Prediction of MHC Class I-Binding Peptides by Dynamic Experiment Design Based on Query Learning with Hidden Markov Models
This item has the following additional materials available:
List of the peptide binding data to a mouse MHC class I molecule Db. Peptide sequences in one-letter code and their normalized log Kd values are presented. The lists are provided as Microsoft EXCEL files.
Excel File - Supplement 1.
List of peptides used for training of HMMs. The initial peptides include binder and non-binder peptides that were available in our laboratory. Peptides listed on the sheets NA, NB, NC and ND are of random source. Peptides used for active training with the Qbag algorithm are listed on the sheets from NE through NK which correspond to the rounds 1 through 7 of the learning cycles.Excel File - Supplement 2.
List of peptides used for the performance evaluation of HMMs.
The Db binding data are presented along with the Lib and Que scores. Random peptides are those used for Fig. 5 and the GenBank peptides are those used for Fig. 6.
Prepared by: the Data Supplement Manager
