Work

PARE: A framework for removal of confounding effects from any distance-based dimension reduction method

Year: 2024

Type: article

Abstract: Dimension reduction tools preserving similarity and graph structure such as t-SNE and UMAP can capture complex biological patterns in high-dimensional data. However, these tools typically are not designed to separate effects of interest from unwanted effects due to confounders. We introduce the partial embedding (PARE) framework, which enables removal of confounders from any distance-based dimension reduction method. We then develop partial t-SNE and partial UMAP and apply these methods to genomic and neuroimaging data. For lower-dimensional visualization, our results show that the PARE framework can remove batch effects in single-cell sequencing data as well as separate clinical and technical variability in neuroimaging measures. We demonstrate that the PARE framework extends dimension reduction methods to highlight biological patterns of interest while effectively removing confounding effects.

Source: PLoS Computational Biology

DOI: 10.1371/journal.pcbi.1012241

Authors: Andrew A. Chen, Kelly Clark, Blake E. Dewey, Anna DuVal, Nicole Pellegrini, Govind Nair, Youmna Jalkh, Samar Khalil, Jon Zurawski, Peter A. Calabresi, Daniel S. Reich, Rohit Bakshi, Haochang Shou, Russell T. Shinohara, Alzheimer’s Disease Neuroimaging Initiative, and North American Imaging in Multiple Sclerosis Cooperative

Institutions: Medical University of South Carolina, University of Pennsylvania, Johns Hopkins University, Johns Hopkins Medicine, National Institutes of Health, National Institute of Neurological Disorders and Stroke, Brigham and Women's Hospital, Harvard University

Language: English

Topic: Single-cell and spatial transcriptomics

Subfield: Molecular Biology

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

Open Access status: gold

Estimated APC paid: $2,926

Funders: National Multiple Sclerosis Society, Bristol-Myers Squibb, Sanofi, University of Pennsylvania, Vertex Pharmaceuticals, Multiple Sclerosis Society, EMD Serono, National Institute of Mental Health, National Institute of Neurological Disorders and Stroke

Awards: G1722533592, G1898893588, Statistical methods for large and complex databases of ultra-high-dimensional, G8351931241, Inter-modal Coupling Image Analytics, Harmonization of Multi-Site Neuroimaging Data from Complex Study Designs, Statistical Methods for Multilevel Multivariate Functional Studies, G970350945

OpenAlex ID: W4400487607

FWCI: 0.271

Cites: 43

Cited by: 2

Related to: 10