Machine learning approaches for prediction of linear B‐cell epitopes on proteins

@article{Sllner2006MachineLA,
  title={Machine learning approaches for prediction of linear B‐cell epitopes on proteins},
  author={Johannes S{\"o}llner and Bernd Mayer},
  journal={Journal of Molecular Recognition},
  year={2006},
  volume={19}
}
Identification and characterization of antigenic determinants on proteins has received considerable attention utilizing both, experimental as well as computational methods. For computational routines mostly structural as well as physicochemical parameters have been utilized for predicting the antigenic propensity of protein sites. However, the performance of computational routines has been low when compared to experimental alternatives. 
Prediction of linear B-cell epitopes
TLDR
This work reviews both classical and novel algorithms and presents its own implementation of the algorithms, using the AAPPred software, for B-cell epitope prediction.
Bioinformatics Resources and Tools for Conformational B-Cell Epitope Prediction
TLDR
The recent advance of bioinformatics resources and tools in conformational B-cell epitope prediction, including databases, algorithms, web servers, and their applications in solving problems in related areas are reviewed.
Recent advances in B-cell epitope prediction methods
TLDR
Recent advances in computational methods for B-cell epitope prediction are reviewed, some gaps in the current state of the art are identified, and some promising directions for improving the reliability of such methods are outlined.
Computational Prediction of Conformational B-Cell Epitopes from Antigen Primary Structures by Ensemble Learning
TLDR
This paper explores various sequence-derived features, which have been observed to be associated with the location of epitopes or ever used in the similar tasks, and develops the ensemble model to predict conformational epitopes from antigen sequences.
Databases for B-cell epitopes.
TLDR
An overview of the important databases associated with the B-cell epitopes is presented and how to compile datasets for the development of B- cell epitope prediction tools is introduced.
SVM-based prediction of linear B-cell epitopes using Bayes Feature Extraction
TLDR
A support vector machines (SVM) prediction model utilizing Bayes Feature Extraction was developed and showed that it was effective in discriminating epitopes from non-epitopes in benchmark datasets and annotated antigenic proteins.
Hybrid methods for B-cell epitope prediction.
  • S. Caoili
  • Biology
    Methods in molecular biology
  • 2014
TLDR
These tasks are considered together herein to clarify their close but often overlooked interrelationship, thereby providing a guide to their performance in mutual support of one another, with emphasis on key physicochemical and biological considerations that are relevant from an applications perspective.
Computational Methods in Linear B-cell Epitope Prediction
TLDR
Various approaches like amino acid scale based methods and machine learning methods used for the prediction of linear B-cell epitopes are reviewed, finding that computational methods for prediction are desirable.
On Predicting Conformational B-cell Epitopes: a Comparative Study and a New Model
TLDR
A novel computa- tional method "CBCPRED" to predict conformational B-cell epitope residues from the target antigen structure by combin- ing support vector machine model with protein structural features and the propensity scores of amino acid physico- chem- ical properties is developed.
Prediction of linear B-cell epitopes using amino acid pair antigenicity scale
TLDR
It is demonstrated that, using SVM (support vector machine) classifier, the AAP antigenicity scale approach has much better performance than the existing scales based on the single amino acid propensity, which is the essence why the new approach is superior to the existing ones.
...
...

References

SHOWING 1-10 OF 26 REFERENCES
BEPITOPE: predicting the location of continuous epitopes and patterns in proteins
TLDR
New functions implemented in the program BEPITOPE to predict continuous protein epitopes include the treatment of a whole genome, the search for a user‐defined pattern, and the combination of prediction to pattern profiles.
Benchmarking B cell epitope prediction: Underperformance of existing methods
TLDR
This work assesses 484 amino acid propensity scales in combination with ranges of plotting parameters to examine exhaustively the correlation of peaks and epitope location within 50 proteins mapped for polyclonal responses and confirms the null hypothesis.
Which structural features determine protein antigenicity
Bcipep: A database of B-cell epitopes
TLDR
A comprehensive database of B-cell epitopes called Bcipep has been developed that covers information on epitopes from a wide range of pathogens and will be source of information for investigators involved in peptide-based vaccine design, disease diagnosis and research in allergy.
Correlation between segmental mobility and the location of antigenic determinants in proteins
Most continuous antigenic determinants of tobacco mosaic virus protein (TMVP), myoglobin and lysozyme correspond to those surface regions in the protein structure, as determined by X-ray
Prediction of protein antigenic determinants from amino acid sequences.
  • T. Hopp, K. R. Woods
  • Biology, Medicine
    Proceedings of the National Academy of Sciences of the United States of America
  • 1981
TLDR
The method was developed using 12 proteins for which extensive immunochemical analysis has been carried out and subsequently was used to predict antigenic determinants for the following proteins, finding that the prediction success rate depended on averaging group length.
...
...