Putting immunoinformatics to the test

NEWS AND VIEWSPutting immunoinformatics to the testLeonard Moise & Anne S De GrootA consensus epitope prediction approach identifies the epitopes responsible for 95% of the murine T-cell response

to vaccinia virus..com/naturebiotechnology Nature Publishing Group Although invasive pathogens may express

thousands of proteins containing millions

of potential epitopes, only a small subset of

these sequences actually stimulates immunity.

Indeed, accurate prediction of epitopes is akin

to finding a needle in a haystack. In this issue,

Moutaftsi et al.1 convincingly demonstrate that

epitope-mapping tools accelerate the discovery

of viral sequences that account for an overwhelmingly large proportion of the response

to vaccinia virus infection in a mouse model.

In addition to revealing the power of immunoinformatics, this study shows that when a host

mounts an anti-viral attack, it may sample and

respond to a broader array of antigens than was

previously thought.Epitope-mapping algorithms, of which

several are available, are relevant not only

to vaccine design but also for characterizing and modifying immune responses in the

context of autoimmunity2, endocrinology3,

allergy4, transplantation5,6, diagnostics7 and

the engineering of therapeutic proteins8. But

controversy has centered around whether

bioinformatics can identify all of the epitopes responsible for protective immunity.

Previous studies demonstrated that a substantial percentage of predicted epitopes

indeed function as such and that a substantial

fraction of known epitopes can be predicted.

However, the fraction of the total immunological response that can be attributed to the

predicted epitopes of a complex virus has yet

to be estimated. Exposure to foreign antigens induces the

adaptive immune system to eliminate infection

and develop protective immunity. Central to

this process are T lymphocytes, which activate

the growth and differentiation of B lymphocytes

and trigger cell-mediated immunity. T cells

perform these roles in response to peptide

fragments of pathogen proteins that are

displayed by major histocompatibility complex

(MHC) proteins at the surface of antigenpresenting cells. Peptides that are presented to

T helper cells are presented by MHC class II

molecules; peptides that trigger cytotoxic

T cell responses are presented by MHC class I

molecules.As the strength of peptide binding to MHC

molecules is a critical determinant of immunogenicity9, algorithms that accurately model the

MHC-peptide interface are central to a priori

strategies for the prediction of T-cell epitopes.

With the availability of complete genome