Dataset Information

Strength in numbers: achieving greater accuracy in MHC-I binding prediction by combining the results from multiple prediction tools.

ABSTRACT: BACKGROUND: Peptides derived from endogenous antigens can bind to MHC class I molecules. Those which bind with high affinity can invoke a CD8+ immune response, resulting in the destruction of infected cells. Much work in immunoinformatics has involved the algorithmic prediction of peptide binding affinity to various MHC-I alleles. A number of tools for MHC-I binding prediction have been developed, many of which are available on the web. RESULTS: We hypothesize that peptides predicted by a number of tools are more likely to bind than those predicted by just one tool, and that the likelihood of a particular peptide being a binder is related to the number of tools that predict it, as well as the accuracy of those tools. To this end, we have built and tested a heuristic-based method of making MHC-binding predictions by combining the results from multiple tools. The predictive performance of each individual tool is first ascertained. These performance data are used to derive weights such that the predictions of tools with better accuracy are given greater credence. The combined tool was evaluated using ten-fold cross-validation and was found to significantly outperform the individual tools when a high specificity threshold is used. It performs comparably well to the best-performing individual tools at lower specificity thresholds. Finally, it also outperforms the combination of the tools resulting from linear discriminant analysis. CONCLUSION: A heuristic-based method of combining the results of the individual tools better facilitates the scanning of large proteomes for potential epitopes, yielding more actual high-affinity binders while reporting very few false positives.

SUBMITTER: Trost B

PROVIDER: S-EPMC1847428 | biostudies-other | 2007

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Strength in numbers: achieving greater accuracy in MHC-I binding prediction by combining the results from multiple prediction tools.

Trost Brett B Bickis Mik M Kusalik Anthony A

Immunome research 20070324

<h4>Background</h4>Peptides derived from endogenous antigens can bind to MHC class I molecules. Those which bind with high affinity can invoke a CD8+ immune response, resulting in the destruction of infected cells. Much work in immunoinformatics has involved the algorithmic prediction of peptide binding affinity to various MHC-I alleles. A number of tools for MHC-I binding prediction have been developed, many of which are available on the web.<h4>Results</h4>We hypothesize that peptides predicte ...[more]

PMID: 17381846

Similar Datasets

Project description:BackgroundMeasures of skeletal muscle function decline at a faster rate with ageing than do indices of skeletal muscle mass. These observations have been attributed to age-related changes in muscle quality, another functional determinant separate from skeletal muscle mass. This study tested the hypothesis that improved predictions of skeletal muscle strength can be accomplished by combining clinically available measures of skeletal muscle mass and quality.MethodsThe participants included 146 healthy adult (age ≥ 18 years, range 18-77 years; X ± SD 47 ± 17 years and body mass index 16.5-51.8 kg/m2 ; 27.7 ± 6.2 kg/m2 ) men (n = 60) and women (n = 86) in whom skeletal muscle mass was estimated as appendicular lean soft tissue (LST) measured by dual-energy X-ray absorptiometry and skeletal muscle quality as bioimpedance analysis-derived phase angle and B-mode-evaluated echogenicity of mid-thigh skeletal muscle. Strength of the right leg and both arms was quantified as knee isokinetic extension and handgrip strength using dynamometers. The statistical significance of adding phase angle or echogenicity to strength prediction multiple regression models that included extremity-specific LST and other covariates (e.g. age and sex) was evaluated to test the study hypothesis.ResultsRight leg LST mass alone was significantly (P < 0.0001) correlated with isokinetic right leg strength (R2 = 0.57). The addition of segmental phase angle measured in the right leg at 50 kHz increased the R2 of this model to 0.66 (P < 0.0001); other phase angle frequencies (5 and 250 kHz) did not contribute significantly to these models. Results were similar for both right and left arm handgrip strength prediction models. Adding age and sex as model covariates increased the R2 values of these models further (e.g. right leg strength model R2 increased to 0.71), but phase angle continued to remain a significant (all P < 0.01) predictor of extremity strength. Similarly, when predicting isokinetic right leg strength, mid-thigh skeletal muscle echogenicity added significantly (P < 0.0001) to right leg LST, increasing R2 from 0.57 to 0.64; age was a significant (P < 0.0001) covariate in this model, increasing R2 further to 0.68.ConclusionsThe hypothesis of the current study was confirmed, strongly supporting and extending earlier reports by quantifying the combined independent effects of skeletal muscle mass and quality on lower-body and upper-body measures of strength. These observations provide a clinically available method for future research aimed at optimizing sarcopenia and frailty risk prediction models.

Dataset Information

Strength in numbers: achieving greater accuracy in MHC-I binding prediction by combining the results from multiple prediction tools.

Publications

Strength in numbers: achieving greater accuracy in MHC-I binding prediction by combining the results from multiple prediction tools.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets