Victor L Jong1, Putri W Novianti2, Kit C B Roes3, Marinus J C Eijkemans3. 1. Biostatistics & Research Support, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, 3508 GA, Utrecht, The Netherlands, Viroscience Lab, Erasmus Medical Center Rotterdam, Rotterdam, CE 3015, The Netherlands and. 2. Biostatistics & Research Support, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, 3508 GA, Utrecht, The Netherlands, Epidemiology & Biostatistics Department, Vrije University Medical Center Amsterdam, HV Amsterdam 1081, The Netherlands. 3. Biostatistics & Research Support, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, 3508 GA, Utrecht, The Netherlands.
Abstract
MOTIVATION: Class predicting with gene expression is widely used to generate diagnostic and/or prognostic models. The literature reveals that classification functions perform differently across gene expression datasets. The question, which classification function should be used for a given dataset remains to be answered. In this study, a predictive model for choosing an optimal function for class prediction on a given dataset was devised. RESULTS: To achieve this, gene expression data were simulated for different values of gene-pairs correlations, sample size, genes' variances, deferentially expressed genes and fold changes. For each simulated dataset, ten classifiers were built and evaluated using ten classification functions. The resulting accuracies from 1152 different simulation scenarios by ten classification functions were then modeled using a linear mixed effects regression on the studied data characteristics, yielding a model that predicts the accuracy of the functions on a given data. An application of our model on eight real-life datasets showed positive correlations (0.33-0.82) between the predicted and expected accuracies. CONCLUSION: The here presented predictive model might serve as a guide to choose an optimal classification function among the 10 studied functions, for any given gene expression data. AVAILABILITY AND IMPLEMENTATION: The R source code for the analysis and an R-package 'SPreFuGED' are available at Bioinformatics online. CONTACT: v.l.jong@umcutecht.nl SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
MOTIVATION: Class predicting with gene expression is widely used to generate diagnostic and/or prognostic models. The literature reveals that classification functions perform differently across gene expression datasets. The question, which classification function should be used for a given dataset remains to be answered. In this study, a predictive model for choosing an optimal function for class prediction on a given dataset was devised. RESULTS: To achieve this, gene expression data were simulated for different values of gene-pairs correlations, sample size, genes' variances, deferentially expressed genes and fold changes. For each simulated dataset, ten classifiers were built and evaluated using ten classification functions. The resulting accuracies from 1152 different simulation scenarios by ten classification functions were then modeled using a linear mixed effects regression on the studied data characteristics, yielding a model that predicts the accuracy of the functions on a given data. An application of our model on eight real-life datasets showed positive correlations (0.33-0.82) between the predicted and expected accuracies. CONCLUSION: The here presented predictive model might serve as a guide to choose an optimal classification function among the 10 studied functions, for any given gene expression data. AVAILABILITY AND IMPLEMENTATION: The R source code for the analysis and an R-package 'SPreFuGED' are available at Bioinformatics online. CONTACT: v.l.jong@umcutecht.nl SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Authors: Amanda D Williams; Olga Y Korolkova; Amos M Sakwe; Timothy M Geiger; Samuel D James; Roberta L Muldoon; Alan J Herline; J Shawn Goodwin; Michael G Izban; Mary K Washington; Duane T Smoot; Billy R Ballard; Maria Gazouli; Amosy E M'Koma Journal: PLoS One Date: 2017-08-17 Impact factor: 3.240
Authors: K L Bowlt Blacklock; Z Birand; L E Selmic; P Nelissen; S Murphy; L Blackwood; J Bass; J McKay; R Fox; S Beaver; M Starkey Journal: Sci Rep Date: 2019-04-24 Impact factor: 4.379
Authors: Kelly Bowlt Blacklock; Zeynep Birand; Deborah Biasoli; Elena Fineberg; Sue Murphy; Debs Flack; Joyce Bass; Stefano Di Palma; Laura Blackwood; Jenny McKay; Trevor Whitbread; Richard Fox; Tom Eve; Stuart Beaver; Mike Starkey Journal: PLoS One Date: 2018-12-19 Impact factor: 3.240