Dataset Information

Algorithms to Improve the Prediction of Postprandial Insulinaemia in Response to Common Foods.

ABSTRACT: Dietary patterns that induce excessive insulin secretion may contribute to worsening insulin resistance and beta-cell dysfunction. Our aim was to generate mathematical algorithms to improve the prediction of postprandial glycaemia and insulinaemia for foods of known nutrient composition, glycemic index (GI) and glycemic load (GL). We used an expanded database of food insulin index (FII) values generated by testing 1000 kJ portions of 147 common foods relative to a reference food in lean, young, healthy volunteers. Simple and multiple linear regression analyses were applied to validate previously generated equations for predicting insulinaemia, and develop improved predictive models. Large differences in insulinaemic responses within and between food groups were evident. GL, GI and available carbohydrate content were the strongest predictors of the FII, explaining 55%, 51% and 47% of variation respectively. Fat, protein and sugar were significant but relatively weak predictors, accounting for only 31%, 7% and 13% of the variation respectively. Nutritional composition alone explained only 50% of variability. The best algorithm included a measure of glycemic response, sugar and protein content and explained 78% of variation. Knowledge of the GI or glycaemic response to 1000 kJ portions together with nutrient composition therefore provides a good approximation for ranking of foods according to their "insulin demand".

SUBMITTER: Bell KJ

PROVIDER: S-EPMC4848679 | biostudies-literature | 2016 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Algorithms to Improve the Prediction of Postprandial Insulinaemia in Response to Common Foods.

Bell Kirstine J KJ Petocz Peter P Colagiuri Stephen S Brand-Miller Jennie C JC

Nutrients 20160408 4

Dietary patterns that induce excessive insulin secretion may contribute to worsening insulin resistance and beta-cell dysfunction. Our aim was to generate mathematical algorithms to improve the prediction of postprandial glycaemia and insulinaemia for foods of known nutrient composition, glycemic index (GI) and glycemic load (GL). We used an expanded database of food insulin index (FII) values generated by testing 1000 kJ portions of 147 common foods relative to a reference food in lean, young, ...[more]

PMID: 27070641

Similar Datasets

Project description:Background: Neoadjuvant chemotherapy is increasingly being used to preoperatively shrink breast tumours prior to surgery. This approach also provides the opportunity to study the molecular changes associated with treatment and evaluate whether on-treatment sequential samples can improve response and outcome predictions over diagnostic or excision samples alone. Methods: A total of 95 samples from a cohort of 50 neoadjuvant chemotherapy-treated primary breast cancer patients (aged 29-76, 48% ER+, 20% HER2+) enrolled in the NEO trial taken before, at 2 weeks on-treatment, mid-therapy and at resection were sequenced with Ion Ampliseq transcriptome yielding expression values for 12,635 genes. Differential expression analysis was performed across 16 responders and 34 non-responders defined by pathological complete response and over treatment time to identify significantly differentially expressed genes, pathways and markers indicative of response status. Prediction accuracy was compared with estimations of established gene signatures, for this dataset and validated using data from the I-SPY1 trial. Results: AAGAB was identified as a novel on-treatment biomarker for pathological complete response, with an accuracy of 100% in the NEO training dataset and 78% accuracy in the I-SPY1 Trial. AAGAB levels on treatment were also significantly predictive of term survival (p = 0.048, p = 0.031) in the two cohorts. This single gene on-treatment biomarker, had greater predictive accuracy than established prognostic tests, Mammaprint and Pam50 risk of recurrence score, although interesting both of these tests performed better in the on-treatment rather than the accepted pre-treatment setting (accuracy improving consistently by 2-8%). Conclusion: Changes in gene expression measured in sequential samples from breast cancer patients receiving neoadjuvant chemotherapy resulted in the identification of a novel on-treatment biomarker and suggest that established prognostic tests may have greater prediction accuracy on- than before treatment. These results support the potential use and further evaluation of on- treatment testing in breast cancer to improve the accuracy of tumour response prediction.

Project description:Pearson's correlation coefficient (ρ) is the most commonly reported metric of the success of prediction in genomic selection (GS). However, in real breeding ρ may not be very useful for assessing the quality of the regression in the tails of the distribution, where individuals are chosen for selection. This research used 14 maize and 16 wheat data sets with different trait-environment combinations. Six different models were evaluated by means of a cross-validation scheme (50 random partitions each, with 90% of the individuals in the training set and 10% in the testing set). The predictive accuracy of these algorithms for selecting individuals belonging to the best α=10, 15, 20, 25, 30, 35, 40% of the distribution was estimated using Cohen's kappa coefficient (κ) and an ad hoc measure, which we call relative efficiency (RE), which indicates the expected genetic gain due to selection when individuals are selected based on GS exclusively. We put special emphasis on the analysis for α=15%, because it is a percentile commonly used in plant breeding programmes (for example, at CIMMYT). We also used ρ as a criterion for overall success. The algorithms used were: Bayesian LASSO (BL), Ridge Regression (RR), Reproducing Kernel Hilbert Spaces (RHKS), Random Forest Regression (RFR), and Support Vector Regression (SVR) with linear (lin) and Gaussian kernels (rbf). The performance of regression methods for selecting the best individuals was compared with that of three supervised classification algorithms: Random Forest Classification (RFC) and Support Vector Classification (SVC) with linear (lin) and Gaussian (rbf) kernels. Classification methods were evaluated using the same cross-validation scheme but with the response vector of the original training sets dichotomised using a given threshold. For α=15%, SVC-lin presented the highest κ coefficients in 13 of the 14 maize data sets, with best values ranging from 0.131 to 0.722 (statistically significant in 9 data sets) and the best RE in the same 13 data sets, with values ranging from 0.393 to 0.948 (statistically significant in 12 data sets). RR produced the best mean for both κ and RE in one data set (0.148 and 0.381, respectively). Regarding the wheat data sets, SVC-lin presented the best κ in 12 of the 16 data sets, with outcomes ranging from 0.280 to 0.580 (statistically significant in 4 data sets) and the best RE in 9 data sets ranging from 0.484 to 0.821 (statistically significant in 5 data sets). SVC-rbf (0.235), RR (0.265) and RHKS (0.422) gave the best κ in one data set each, while RHKS and BL tied for the last one (0.234). Finally, BL presented the best RE in two data sets (0.738 and 0.750), RFR (0.636) and SVC-rbf (0.617) in one and RHKS in the remaining three (0.502, 0.458 and 0.586). The difference between the performance of SVC-lin and that of the rest of the models was not so pronounced at higher percentiles of the distribution. The behaviour of regression and classification algorithms varied markedly when selection was done at different thresholds, that is, κ and RE for each algorithm depended strongly on the selection percentile. Based on the results, we propose classification method as a promising alternative for GS in plant breeding.

Dataset Information

Algorithms to Improve the Prediction of Postprandial Insulinaemia in Response to Common Foods.

Publications

Algorithms to Improve the Prediction of Postprandial Insulinaemia in Response to Common Foods.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets