Data denoising with transfer learning in single-cell transcriptomics

Nat Methods. 2019 Sep;16(9):875-878. doi: 10.1038/s41592-019-0537-1. Epub 2019 Aug 30.

Abstract

Single-cell RNA sequencing (scRNA-seq) data are noisy and sparse. Here, we show that transfer learning across datasets remarkably improves data quality. By coupling a deep autoencoder with a Bayesian model, SAVER-X extracts transferable gene-gene relationships across data from different labs, varying conditions and divergent species, to denoise new target datasets.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Animals
  • Bayes Theorem
  • Breast Neoplasms / metabolism*
  • Computational Biology / methods*
  • Female
  • Gene Expression Profiling
  • Gene Expression Regulation
  • Humans
  • Leukocytes, Mononuclear / metabolism*
  • Mice
  • Sequence Analysis, RNA / methods
  • Sequence Analysis, RNA / standards*
  • Single-Cell Analysis / methods*
  • T-Lymphocytes / metabolism*
  • Transcriptome*