An efficient visualisation method for exploring latent patterns in large microbiome expression data sets
by Weiwei Xu; Timothy Schultz; Rong Xie
International Journal of Data Mining and Bioinformatics (IJDMB), Vol. 15, No. 1, 2016

Abstract: In recent years, HMP has provided analytical insights into the human microbiome to gain better insights into its effects on human health. Applying insights for therapeutic and biotechnological applications requires researchers efficiently classify the vast amount of microbes which comprise the microbiome. Since these datasets are sparse and complex in nature, the application of dimensionality reduction algorithms is a popular way for extracting latent phylogenetic themes. We introduce an Augmented Barnes-Hut t-SNE method, which is both more efficient in processing time and sensitive to subtle albeit meaningful variations in microbial classifications based on 5 high-level anatomical regions. We demonstrate that our method not only separates the microbiome into these regions, but can further elucidate them into 18 separate sample sites. It is contended that this approach can accurately resolve phylogenetic themes at varying levels of granularity, and anticipate its application in other research domains where complex high-dimensional datasets are prevalent.

Online publication date: Thu, 21-Apr-2016

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Data Mining and Bioinformatics (IJDMB):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com