ReviewCOMPUTATIONAL BIOLOGY

Avoiding common pitfalls when clustering biological data

See allHide authors and affiliations

Sci. Signal.  14 Jun 2016:
Vol. 9, Issue 432, pp. re6
DOI: 10.1126/scisignal.aad1932

You are currently viewing the abstract.

View Full Text

Gloss

Clustering is an unsupervised learning method, grouping data points based on similarity, with the goal of revealing the underlying structure of data. Advances in molecular biology have yielded large and complex data sets, making clustering essential to understand and visualize the data. Clustering can be a powerful technique, but it harbors potential pitfalls due to the high-dimensional nature of biological data, the failure to consider more than one clustering method for a given problem, and the difficulty in determining whether clustering has produced meaningful results. We present concrete examples of problems and solutions (clustering results) in the form of toy problems and real biological data for these pitfalls, illustrating how to avoid overinterpreting the data and missing valuable insights within high-throughput molecular measurements. The article contains six figures, four tables, and 77 references.

Abstract

Clustering is an unsupervised learning method, which groups data points based on similarity, and is used to reveal the underlying structure of data. This computational approach is essential to understanding and visualizing the complex data that are acquired in high-throughput multidimensional biological experiments. Clustering enables researchers to make biological inferences for further experiments. Although a powerful technique, inappropriate application can lead biological researchers to waste resources and time in experimental follow-up. We review common pitfalls identified from the published molecular biology literature and present methods to avoid them. Commonly encountered pitfalls relate to the high-dimensional nature of biological data from high-throughput experiments, the failure to consider more than one clustering method for a given problem, and the difficulty in determining whether clustering has produced meaningful results. We present concrete examples of problems and solutions (clustering results) in the form of toy problems and real biological data for these issues. We also discuss ensemble clustering as an easy-to-implement method that enables the exploration of multiple clustering solutions and improves robustness of clustering solutions. Increased awareness of common clustering pitfalls will help researchers avoid overinterpreting or misinterpreting the results and missing valuable insights when clustering biological data.

View Full Text