Online citations, reference lists, and bibliographies.
← Back to Search

Discovering What Dimensionality Reduction Really Tells Us About RNA-Seq Data

Sean Simmons, J. Peng, J. Bienkowska, B. Berger
Published 2015 · Computer Science, Medicine

Save to my Library
Download PDF
Analyze on Scholarcy
Abstract Biology is being inundated by noisy, high-dimensional data to an extent never before experienced. Dimensionality reduction techniques such as principal component analysis (PCA) are common approaches for dealing with this onslaught. Though these unsupervised techniques can help uncover interesting structure in high-dimensional data they give little insight into the biological and technical considerations that might explain the uncovered structure. Here we introduce a hybrid approach—component selection using mutual information (CSUMI)—that uses a mutual information—based statistic to reinterpret the results of PCA in a biologically meaningful way. We apply CSUMI to RNA-seq data from GTEx. Our hybrid approach enables us to unveil the previously hidden relationship between principal components (PCs) and the underlying biological and technical sources of variation across samples. In particular, we look at how tissue type affects PCs beyond the first two, allowing us to devise a principled way of choosing which PCs to consider when exploring the data. We further apply our method to RNA-seq data taken from the brain and show that some of the most biologically informative PCs are higher-dimensional PCs; for instance, PC 5 can differentiate the basal ganglia from other tissues. We also use CSUMI to explore how technical artifacts affect the global structure of the data, validating previous results and demonstrating how our method can be viewed as a verification framework for detecting undiscovered biases in emerging technologies. Finally we compare CSUMI to two correlation-based approaches, showing ours outperforms both. A python implementation is available online on the CSUMI website.
This paper references
Genome-wide identification of post-translational modulators of transcription factor activity
K 763–774. Wang (2009)
High-Resolution Modeling of Cellular Signaling Networks
Michael Baym (2008)
High-resolution modeling of cellular signaling networks. RECOMB. 4955
M Baym (2008)
A gene expression profile of stem cell pluripotentiality and differentiation is conserved across diverse solid and hematopoietic cancers
N. Palmer (2012)
Estimating mutual information using B-spline functions – an improved similarity measure for analysing gene expression data
C. Daub (2003)
Feature selection based on mutual information: criteria of max-dependency
F. Long (2005)
Cell type–specific gene expression differences in complex tissues
S. Shen-Orr (2010)
Principal component analysis for clustering gene expression data
K. Yeung (2001)
Synthetic maps of human gene frequencies in Europeans.
P. Menozzi (1978)
IsoRankN: spectral methods for global alignment of multiple protein networks
Chung-Shou Liao (2009)
A comprehensive assessment of RNA-seq accuracy, reproducibility and information content by the Sequencing Quality Control consortium
Z. Su (2014)
Principal components analysis for stratification in genome-wide association studies
A Price (2006)
On Estimating P Values by Monte Carlo Methods
W. Ewens (2003)
Detecting Novel Associations in Large Data Sets
David N. Reshef (2011)
E-mail: bab@mit
Preserving biological heterogeneity with a permuted surrogate variable analysis for genomics batch correction
H. Parker (2014)
High-resolution modeling of cellular signaling
M. Baym (2008)
Single-Cell Trajectory Detection Uncovers Progression and Regulatory Coordination in Human B Cell Development
S. Bendall (2014)
Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy
H. Peng (2005)
A gene expression profile of stem cell pluripotentiality and differentiation is conserved across
N Palmer (2012)
Address correspondence to: Dr. Bonnie Berger Department of Mathematics Massachusetts Institute of Technology 77 Mass Avenue
A note on calculation of empirical P values from Monte Carlo procedure.
B. North (2003)
On the distribution of the largest eigenvalues in principal component analysis
I. Johnstone (2001)
max-relevance, and min-redundancy
Making sense out of massive data by going beyond differential expression
Patrick R. Schmid (2012)
diverse solid and hematopoietic cancers
Defining diversity, specialization, and gene specificity in transcriptomes through information theory
O. Martinez (2008)
Genome-wide Identification of Post-translational Modulators of Transcription Factor Activity in Human B-Cells
K. Wang (2009)
Nonnegative Matrix Factorization: An Analytical and Interpretive Tool in Computational Biology
K. Devarajan (2008)
Network-based stratification of tumor mutations
Matan Hofree (2013)
The Genotype-Tissue Expression (GTEx) project
J. Lonsdale (2013)

This paper is referenced by
Semantic Scholar Logo Some data provided by SemanticScholar