A1 Refereed original research article in a scientific journal
qSNE: quadratic rate t-SNE optimizer with automatic parameter tuning for large datasets
Authors: Häkkinen A, Koiranen J, Casado J, Kaipio K, Lehtonen O, Petrucci E, Hynninen J, Hietanen S, Carpén O, Pasquini L, Biffoni M, Lehtonen R, Hautaniemi S
Publication year: 2020
Journal: Bioinformatics
Journal name in source: Bioinformatics (Oxford, England)
Journal acronym: Bioinformatics
Volume: 36
Issue: 20
First page : 5086
Last page: 5092
Number of pages: 7
ISSN: 1367-4803
eISSN: 1367-4811
DOI: https://doi.org/10.1093/bioinformatics/btaa637
Self-archived copy’s web address: https://research.utu.fi/converis/portal/Publication/50793942
Motivation: Non-parametric dimensionality reduction techniques, such as t-distributed stochastic neighbor embedding (t-SNE), are the most frequently used methods in the exploratory analysis of single-cell datasets. Current implementations scale poorly to massive datasets and often require downsampling or interpolative approximations, which can leave less-frequent populations undiscovered and much information unexploited.
Results: We implemented a fast t-SNE package, qSNE, which uses a quasi-Newton optimizer, allowing quadratic convergence rate and automatic perplexity (level of detail) optimizer. Our results show that these improvements make qSNE significantly faster than regular t-SNE packages and enables full analysis of large datasets, such as mass cytometry data, without downsampling.
Downloadable publication This is an electronic reprint of the original article. |