Paper Info
Title
Novel machine learning methods for MHC Class I binding prediction
Abstract
MHC class I molecules are key players in the human immune system. They bind small peptides derived from intracellular proteins and present them on the cell surface for surveillance by the immune system. Prediction of such MHC class I binding peptides is a vital step in the design of peptide-based vaccines and therefore one of the major problems in computational immunology. Thousands of different types of MHC class I molecules exist, each displaying a distinct binding specificity. The lack of sufficient training data for the majority of these molecules hinders the application of Machine Learning to this problem. We propose two approaches to improve the predictive power of kernel-based Machine Learning methods for MHC class I binding prediction: First, a modification of the Weighted Degree string kernel that allows for the incorporation of amino acid properties. Second, we propose an enhanced Multitask kernel and an optimization procedure to fine-tune the kernel parameters. The combination of both approaches yields improved performance, which we demonstrate on the IEDB benchmark data set.
Year
DOI
Venue
2010
10.1007/978-3-642-16001-1_9
PRIB
Keywords
Field
DocType
binding peptides,iedb benchmark data,mhc class i molecule,novel machine,kernel parameter,machine learning,enhanced multitask kernel,weighted degree string kernel,mhc class,distinct binding specificity,binding prediction,string kernel,mhc class i,human immune system,immune system
Training set,Kernel (linear algebra),Computer science,MHC class I,Computational immunology,Artificial intelligence,Bioinformatics,Binding selectivity,String kernel,Machine learning
Conference
Volume
ISSN
ISBN
6282
0302-9743
3-642-16000-X
Citations 
PageRank 
References 
8
0.44
15
Authors
5
Name
Order
Citations
PageRank
Christian Widmer123927.29
Nora C Toussaint21107.40
yasemin altun32463150.46
Oliver Kohlbacher4975101.91
Gunnar Rätsch55625671.20