Spectral Clustering Strategies for Heterogeneous Disease Expression Data

Abstract

Clustering of gene expression data simplifies subsequent data analyses and forms the basis of numerous approaches for biomarker identification, prediction of clinical outcome, and personalized therapeutic strategies. The most popular clustering methods such as K-means and hierarchical clustering are intuitive and easy to use, but they require arbitrary choices on their various parameters (number of clusters for K-means, and a threshold to cut the tree for hierarchical clustering). Human disease gene expression data are in general more difficult to cluster efficiently due to background (genotype) heterogeneity, disease stage and progression differences and disease subtyping; all of which cause gene expression datasets to be more heterogeneous.Spectral clustering has been recently introduced in many fields as a promising alternative to standard clustering methods. The idea is that pairwise comparisons can help reveal global features through the eigen techniques. In this paper, we developed a new recursive K-means spectral clustering method (ReKS) for disease gene expression data. We benchmarked ReKS on three large-scale cancer datasets and we compared it to different clustering methods with respect to execution time, background models and external biological knowledge. We found ReKS to be superior to the hierarchical methods and equally good to K-means, but much faster than them and without the requirement for a priori knowledge of K. Overall, ReKS offers an attractive alternative for efficient clustering of human disease data.

T-Recs: Stable Selection of Dynamically Formed Groups of Features With Application to Prediction of Clinical Outcomes

“…Therefore, primary feature selection here is equivalent to discovering the PC(T) . For generating the tree structure we use ReKS ( Recursive K-means Spectral Clustering ), which was shown to outperform other methods in terms of speed or efficiency and outputs more balanced trees when applied to heterogeneous clinical data (10). Finally, to create the representative features of a cluster we tested the first Principal Component of the cluster, the medoid, and the centroid of the clustered variables.…”

“…The complexity of T-ReCS is roughly O(|φ | 2 ). Specifically, ReKS is O(|φ | 2 ) (10), MMPC is O(|φ |•| PC(T) |• k ), and conditional independence tests for ascending the tree is O( log |φ |•| PC(T) |). We note, however, that selection of different methods for single feature selection and tree construction can alter this complexity.…”

Image Clustering Method based on Particle Swarm Optimization

“…Another effective clustering method was proposed for heterogeneous disease expression data [17]. Recursive Kmeans spectral clustering method (ReKS) was developed, which was found to be superior to the hierarchical clustering method and much faster than k-means.…”

Measuring Patient Similarities via a Deep Architecture with Medical Concept Embedding

“…To capture underlying structure in the history of present illness section from patients EHR, Henao [14] proposed a statistical model that groups patients based on text data in the initial history of present illness (HPI) and final diagnosis (DX) of a patients EHR. For human disease gene expression, Huang [15] presented a new recursive Kmeans spectral clustering method (ReKS) to efficient cluster human diseases. Most of these research have demonstrate effectiveness of their model with real-world experiments, that convinces us of the applicability of clustering patients on cohorts discovering.…”

“…The parameters of our deep learning were as follow: the width of the convolution filters w is set to 5,10,15,20,25, and the number of convolutional feature maps m takes on 50, 100, 150, 200. We use stochastic gradient descent to optimize the model's parameters.…”