Abstract
BackgroundA feature selection method in microarray gene expression data should be independent of platform, disease and dataset size. Our hypothesis is that among the statistically significant ranked genes in a gene list, there should be clusters of genes that share similar biological functions related to the investigated disease. Thus, instead of keeping N top ranked genes, it would be more appropriate to define and keep a number of gene cluster exemplars.ResultsWe propose a hybrid FS method (mAP-KL), which combines multiple hypothesis testing and affinity propagation (AP)-clustering algorithm along with the Krzanowski & Lai cluster quality index, to select a small yet informative subset of genes. We applied mAP-KL on real microarray data, as well as on simulated data, and compared its performance against 13 other feature selection approaches. Across a variety of diseases and number of samples, mAP-KL presents competitive classification results, particularly in neuromuscular diseases, where its overall AUC score was 0.91. Furthermore, mAP-KL generates concise yet biologically relevant and informative N-gene expression signatures, which can serve as a valuable tool for diagnostic and prognostic purposes, as well as a source of potential disease biomarkers in a broad range of diseases.ConclusionsmAP-KL is a data-driven and classifier-independent hybrid feature selection method, which applies to any disease classification problem based on microarray data, regardless of the available samples. Combining multiple hypothesis testing and AP leads to subsets of genes, which classify unknown samples from both, small and large patient cohorts with high accuracy.
Highlights
A feature selection method in microarray gene expression data should be independent of platform, disease and dataset size
Rationale for selecting the proposed approach Jaeger et al [16] claimed that ranking algorithms produce lists of genes, where the top ranked genes are highly correlated with each other, mainly because they belong to the same pathway
Regarding the number of genes, we employ a clustering index to determine the ‘actual’ number of representative genes. This differs from mRMR method, which iterates in its ranked gene list before concluding to a subset, and from Jaeger and Hanczar, where the resultant subset is driven by the initial number of potential clusters, which is set arbitrarily
Summary
A feature selection method in microarray gene expression data should be independent of platform, disease and dataset size. Informative genes are selected according to a two-sample statistical test combined with multiple testing procedures to guard against Type 1 errors [1]. A wide variety of FS algorithms has been proposed [3,4,5] and depending on how they combine the feature selection search with the construction of the classification model they can be classified into 3 categories: filter, wrapper, and embedded [2]. Embedded algorithms, like Random Forest [14], select the best subset of genes incorporating the classifiers’ bias [2]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
