bioRxivNeurosciencePreprint
Deconfounded Dimension Reduction via Partial Embeddings
Andrew A. CheniD, Kelly A. ClarkiD, Blake E. DeweyiD, Anna L. DuValiD, Nicole PellegriniiD, Govind NairiD, Youmna Jalkh, Samar Khalil, Jon Zurawski, Peter A. CalabresiiD, Daniel S. ReichiD, Rohit BakshiiD, Haochang ShouiD, Russell Taki ShinoharaiD, the Alzheimer’s Disease Neuroimaging Initiative, the North American Imaging in Multiple Sclerosis Cooperative
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. 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.
- assay: bulk RNA-seq
- assay: imaging
- research focus: batch correction / integration
- research focus: dimensionality reduction
- approach: statistical model
- Citations
- 1
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Embedding to Reference t-SNE Space Addresses Batch Effects in Single-Cell Classification
Pavlin G. Poličar, Martin Stražar, Blaž Zupan·Machine LearningPeer-reviewed·Aug 24, 2021
- Exploring dimension-reduced embeddings with Sleepwalk
Svetlana Ovchinnikova, Simon Anders·Genome ResearchPeer-reviewed·May 1, 2020
- EMBEDR: Distinguishing Signal from Noise in Single-Cell Omics Data
Eric M. Johnson, William L. Kath, Madhav Mani·PatternsPeer-reviewed·Jan 1, 2021
- projectR: An R/Bioconductor package for transfer learning via PCA, NMF, correlation, and clustering
Gaurav Sharma, Carlo Colantuoni, … Genevieve Stein-O’Brien·BioinformaticsPeer-reviewed·Mar 11, 2020
- Faithful Supervised Dimensionality Reduction for Biomedical Data via Decision Geometry
Zexuan Wang, Zhuoping Zhou, … Li Shen·ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB)Peer-reviewed·Jun 30, 2026
- NewWave: a scalable R/Bioconductor package for the dimensionality reduction and batch effect removal of single-cell RNA-seq data
Federico Agostinis, Chiara Romualdi, … Davide Risso·BioinformaticsPeer-reviewed·Mar 8, 2022