학술논문
Metric multidimensional scaling for large single-cell datasets using neural networks.
Document Type
Academic Journal
Author
Canzar S; Faculty of Informatics and Data Science, University of Regensburg, Regensburg, Germany. stefan.canzar@ur.de.; Do VH; Center for Applied Mathematics and Informatics, Le Quy Don Technical University, Hanoi, Vietnam.; Jelić S; School of Applied Mathematics and Informatics, University of Osijek, Osijek, Croatia.; Laue S; Department of Informatics, Universität Hamburg, Hamburg, Germany.; Matijević D; School of Applied Mathematics and Informatics, University of Osijek, Osijek, Croatia.; Prusina T; Department of Informatics, Universität Hamburg, Hamburg, Germany.
Source
Publisher: BioMed Central Country of Publication: England NLM ID: 101265088 Publication Model: Electronic Cited Medium: Print ISSN: 1748-7188 (Print) Linking ISSN: 17487188 NLM ISO Abbreviation: Algorithms Mol Biol Subsets: PubMed not MEDLINE
Subject
Language
English
ISSN
1748-7188
Abstract
Metric multidimensional scaling is one of the classical methods for embedding data into low-dimensional Euclidean space. It creates the low-dimensional embedding by approximately preserving the pairwise distances between the input points. However, current state-of-the-art approaches only scale to a few thousand data points. For larger data sets such as those occurring in single-cell RNA sequencing experiments, the running time becomes prohibitively large and thus alternative methods such as PCA are widely used instead. Here, we propose a simple neural network-based approach for solving the metric multidimensional scaling problem that is orders of magnitude faster than previous state-of-the-art approaches, and hence scales to data sets with up to a few million cells. At the same time, it provides a non-linear mapping between high- and low-dimensional space that can place previously unseen cells in the same embedding.
(© 2024. The Author(s).)
(© 2024. The Author(s).)