Dataset Information

Misclassification Errors in Unsupervised Classification Methods. Comparison Based on the Simulation of Targeted Proteomics Data.

ABSTRACT: Unsupervised classification methods are gaining acceptance in omics studies of complex common diseases, which are often vaguely defined and are likely the collections of disease subtypes. Unsupervised classification based on the molecular signatures identified in omics studies have the potential to reflect molecular mechanisms of the subtypes of the disease and to lead to more targeted and successful interventions for the identified subtypes. Multiple classification algorithms exist but none is ideal for all types of data. Importantly, there are no established methods to estimate sample size in unsupervised classification (unlike power analysis in hypothesis testing). Therefore, we developed a simulation approach allowing comparison of misclassification errors and estimating the required sample size for a given effect size, number, and correlation matrix of the differentially abundant proteins in targeted proteomics studies. All the experiments were performed in silico. The simulated data imitated the expected one from the study of the plasma of patients with lower urinary tract dysfunction with the aptamer proteomics assay Somascan (SomaLogic Inc, Boulder, CO), which targeted 1129 proteins, including 330 involved in inflammation, 180 in stress response, 80 in aging, etc. Three popular clustering methods (hierarchical, k-means, and k-medoids) were compared. K-means clustering performed much better for the simulated data than the other two methods and enabled classification with misclassification error below 5% in the simulated cohort of 100 patients based on the molecular signatures of 40 differentially abundant proteins (effect size 1.5) from among the 1129-protein panel.

SUBMITTER: Andreev VP

PROVIDER: S-EPMC4982549 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Misclassification Errors in Unsupervised Classification Methods. Comparison Based on the Simulation of Targeted Proteomics Data.

Andreev Victor P VP Gillespie Brenda W BW Helfand Brian T BT Merion Robert M RM

Journal of proteomics & bioinformatics 20160516

Unsupervised classification methods are gaining acceptance in omics studies of complex common diseases, which are often vaguely defined and are likely the collections of disease subtypes. Unsupervised classification based on the molecular signatures identified in omics studies have the potential to reflect molecular mechanisms of the subtypes of the disease and to lead to more targeted and successful interventions for the identified subtypes. Multiple classification algorithms exist but none is ...[more]

PMID: 27524871

Similar Datasets

Project description:BackgroundRecent literature on the comparison of machine learning methods has raised questions about the neutrality, unbiasedness and utility of many comparative studies. Reporting of results on favourable datasets and sampling error in the estimated performance measures based on single samples are thought to be the major sources of bias in such comparisons. Better performance in one or a few instances does not necessarily imply so on an average or on a population level and simulation studies may be a better alternative for objectively comparing the performances of machine learning algorithms.MethodsWe compare the classification performance of a number of important and widely used machine learning algorithms, namely the Random Forests (RF), Support Vector Machines (SVM), Linear Discriminant Analysis (LDA) and k-Nearest Neighbour (kNN). Using massively parallel processing on high-performance supercomputers, we compare the generalisation errors at various combinations of levels of several factors: number of features, training sample size, biological variation, experimental variation, effect size, replication and correlation between features.ResultsFor smaller number of correlated features, number of features not exceeding approximately half the sample size, LDA was found to be the method of choice in terms of average generalisation errors as well as stability (precision) of error estimates. SVM (with RBF kernel) outperforms LDA as well as RF and kNN by a clear margin as the feature set gets larger provided the sample size is not too small (at least 20). The performance of kNN also improves as the number of features grows and outplays that of LDA and RF unless the data variability is too high and/or effect sizes are too small. RF was found to outperform only kNN in some instances where the data are more variable and have smaller effect sizes, in which cases it also provide more stable error estimates than kNN and LDA. Applications to a number of real datasets supported the findings from the simulation study.

Project description:The choice of the most appropriate unsupervised machine-learning method for "heterogeneous" or "mixed" data, i.e. with both continuous and categorical variables, can be challenging. Our aim was to examine the performance of various clustering strategies for mixed data using both simulated and real-life data. We conducted a benchmark analysis of "ready-to-use" tools in R comparing 4 model-based (Kamila algorithm, Latent Class Analysis, Latent Class Model [LCM] and Clustering by Mixture Modeling) and 5 distance/dissimilarity-based (Gower distance or Unsupervised Extra Trees dissimilarity followed by hierarchical clustering or Partitioning Around Medoids, K-prototypes) clustering methods. Clustering performances were assessed by Adjusted Rand Index (ARI) on 1000 generated virtual populations consisting of mixed variables using 7 scenarios with varying population sizes, number of clusters, number of continuous and categorical variables, proportions of relevant (non-noisy) variables and degree of variable relevance (low, mild, high). Clustering methods were then applied on the EPHESUS randomized clinical trial data (a heart failure trial evaluating the effect of eplerenone) allowing to illustrate the differences between different clustering techniques. The simulations revealed the dominance of K-prototypes, Kamila and LCM models over all other methods. Overall, methods using dissimilarity matrices in classical algorithms such as Partitioning Around Medoids and Hierarchical Clustering had a lower ARI compared to model-based methods in all scenarios. When applying clustering methods to a real-life clinical dataset, LCM showed promising results with regard to differences in (1) clinical profiles across clusters, (2) prognostic performance (highest C-index) and (3) identification of patient subgroups with substantial treatment benefit. The present findings suggest key differences in clustering performance between the tested algorithms (limited to tools readily available in R). In most of the tested scenarios, model-based methods (in particular the Kamila and LCM packages) and K-prototypes typically performed best in the setting of heterogeneous data.

Dataset Information

Misclassification Errors in Unsupervised Classification Methods. Comparison Based on the Simulation of Targeted Proteomics Data.

Publications

Misclassification Errors in Unsupervised Classification Methods. Comparison Based on the Simulation of Targeted Proteomics Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets