PARE: A framework for removal of confounding effects from any distance-based dimension reduction met...
PARE: A framework for removal of confounding effects from any distance-based dimension reduction method
About this item
Full title
Author / Creator
Alzheimer’s Disease Neuroimaging Initiative, and North American Imaging in Multiple Sclerosis Cooperative , Chen, Andrew A. , Clark, Kelly , Dewey, Blake E. , DuVal, Anna , Pellegrini, Nicole , Nair, Govind , Jalkh, Youmna , Khalil, Samar , Zurawski, Jon , Calabresi, Peter A. , Reich, Daniel S. , Bakshi, Rohit , Shou, Haochang and Shinohara, Russell T.
Publisher
United States: Public Library of Science
Journal title
Language
English
Formats
Publication information
Publisher
United States: Public Library of Science
Subjects
More information
Scope and Contents
Contents
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 enabl...
Alternative Titles
Full title
PARE: A framework for removal of confounding effects from any distance-based dimension reduction method
Authors, Artists and Contributors
Author / Creator
Chen, Andrew A.
Clark, Kelly
Dewey, Blake E.
DuVal, Anna
Pellegrini, Nicole
Nair, Govind
Jalkh, Youmna
Khalil, Samar
Zurawski, Jon
Calabresi, Peter A.
Reich, Daniel S.
Bakshi, Rohit
Shou, Haochang
Shinohara, Russell T.
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_doaj_primary_oai_doaj_org_article_6dc81d154df04a90a16c0ae4b80ffc44
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_doaj_primary_oai_doaj_org_article_6dc81d154df04a90a16c0ae4b80ffc44
Other Identifiers
ISSN
1553-7358,1553-734X
E-ISSN
1553-7358
DOI
10.1371/journal.pcbi.1012241
