Quantitative online prediction of peptide binding to the major histocompatibility complex

Hattotuwagama, Channa K.; Guan, Peipei; Doytchinova, Irini; Zygouri, Christianna; Flower, Darren R.

doi:10.1016/s1093-3263(03)00160-8

Cited by 61 publications

(47 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This technique was further validated by coupling the in silico predictions to in vitro analysis [19]. Thus the modern tools for designing immunovaccines include QSAR [20] (both 3D-QSAR [21 -23] and 2D methods [24,25]) and molecular dynamic simulations using high-performance computing [17,26]. There are also online tools available for the prediction of affinity and activity of peptides.…”

Section: Introductionmentioning

confidence: 99%

Encoding Type and Position in Peptide QSAR: Application to Peptides Binding to Class I MHC Molecule HLA‐A*0201

et al. 2007

View full text Add to dashboard Cite

Various QSAR approaches have evolved over time to find a relationship between chemical structure and biological response by translating the information in the chemical structure into a quantitative (Hansch approach) or qualitative (Free -Wilson approach) description followed by a statistical modeling. The nature and position of every amino acid in the peptide sequence dictate its binding to the receptor. We have investigated an approach to peptide QSAR where the position and type of every amino acid in the sequence are encoded in the QSAR paradigm, and along with the principles of the Hansch (descriptor QSAR) and the Free -Wilson (binary QSAR) methodologies we were able to arrive at the optimal sequence of amino acids in the peptide. As an example, we illustrate the procedure on two datasets of nonamer peptides that bind to the class I major histocompatibility complex molecule HLA-A*0201. The models generated have statistically significant co-relation coefficient (r 2 ) and predictive r 2 (r 2 pred ). The crossvalidated r 2 (q 2 ) for "descriptor QSAR" approach and the "binary QSAR" approach are in an acceptable range. The QSAR analysis is able to identify the preferential and detrimental amino acids at every position of the nonamer peptide. This QSAR formalism as applied to peptides is a simple and effective way to explain structure -activity relationships as well as to optimize the binding affinity.

show abstract

Section: Introductionmentioning

confidence: 99%

Encoding Type and Position in Peptide QSAR: Application to Peptides Binding to Class I MHC Molecule HLA‐A*0201

et al. 2007

View full text Add to dashboard Cite

show abstract

“…They have been used for predicting MHC class I binding peptides and the results in [15] indicate that they predict well even on relatively small peptide datasets. Multivariate statistical approaches [16] are based on partial least squares. The prediction is obtained from a combination of individual amino acid contributions at each position of the peptide and contributions from side chain side to chain interactions.…”

Section: Review Of Current Machine Learning Methods In Predicting Hla-mentioning

confidence: 99%

Extreme Learning Machine for Predicting HLA-Peptide Binding

Handoko

Keong

Soon

et al. 2006

Advances in Neural Networks - ISNN 2006

View full text Add to dashboard Cite

Abstract. Machine learning techniques have been recognized as powerful tools for learning from data. One of the most popular learning techniques, the BackPropagation (BP) Artificial Neural Networks, can be used as a computer model to predict peptides binding to the Human Leukocyte Antigens (HLA). The major advantage of computational screening is that it reduces the number of wet-lab experiments that need to be performed, significantly reducing the cost and time. A recently developed method, Extreme Learning Machine (ELM), which has superior properties over BP has been investigated to accomplish such tasks. In our work, we found that the ELM is as good as, if not better than, the BP in term of time complexity, accuracy deviations across experiments, andmost importantly -prevention from over-fitting for prediction of peptide binding to HLA.

show abstract

“…We compared the predictive performance of MHCMIR with that of several MHC-II binding peptide prediction methods reported in the literature: Gibbs sampler (Nielsen et al, 2004), TEPITOPE (Sturniolo et al, 1999), SVRMHC (Liu et al, 2006), MHCPred (Hattotuwagama et al, 2004), ARB , NetMCHII (Nielsen et al, 2007), and MOEA (Rajapakse et al, 2007). Because most reports of MHC-II binding activity prediction methods in the literature focus on qualitative prediction of MHC-II binding activity, although MHCMIR is able to produce both quantitative and qualitative predictions of MHC-II binding activity (the latter by comparing the predicted binding affinity value with a threshold), our comparisons focus on qualitative predictions of MHC-II binding activity.…”

Section: Resultsmentioning

confidence: 99%

“…A variety of methods for predicting MHC-I binding peptides from amino acid sequence information have been proposed. Examples of these MHC-I peptide prediction methods include methods based on: scoring matrices (Parker et al, 1994;Rammensee et al, 1999;Reche et al, 2004;Bui et al, 2005;Peters and Sette, 2005); hidden Markov models (HMM) (Mamitsuka, 1998); additive method (Hattotuwagama et al, 2004); artificial neural networks (ANN) ; support vector machine (SVM) (Donnes and Kohlbacher, 2006); support vector regression (SVR) (Liu et al, 2006). These methods can be categorized into two major types: i) qualitative methods (e.g., (Reche et al, 2004;Donnes and Kohlbacher, 2006)), which predict whether a test peptide is an MHC-I binder or non-binder; ii) quantitative methods (e.g., (Hattotuwagama et al, 2004;Liu et al, 2006;Peters and Sette, 2005)), which predicts the value of the binding affinity of a test peptide.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning approaches for epitope prediction

EL-Manzalawy¹

View full text Add to dashboard Cite

Quantitative online prediction of peptide binding to the major histocompatibility complex

Cited by 61 publications

References 33 publications

Encoding Type and Position in Peptide QSAR: Application to Peptides Binding to Class I MHC Molecule HLA‐A*0201

Encoding Type and Position in Peptide QSAR: Application to Peptides Binding to Class I MHC Molecule HLA‐A*0201

Extreme Learning Machine for Predicting HLA-Peptide Binding

Machine learning approaches for epitope prediction

Contact Info

Product

Resources

About