Rxivist logo

Rxivist combines preprints from bioRxiv with data from Twitter to help you find the papers being discussed in your field. Currently indexing 59,758 bioRxiv papers from 265,632 authors.

Imputation of single-cell gene expression with an autoencoder neural network

By Md B. Badsha, Rui Li, Boxiang Liu, Yang I Li, Min Xian, Nicholas E Banovich, Audrey Q. Fu

Posted 29 Dec 2018
bioRxiv DOI: 10.1101/504977

Single-cell RNA-sequencing (scRNA-seq) is a rapidly evolving technology that enables measurement of gene expression levels at an unprecedented resolution. Despite the explosive growth in the number of cells that can be assayed by a single experiment, scRNA-seq still has several limitations, including high rates of dropouts. High dropout rates result in a large number of genes having zero read count in the scRNA-seq data, which complicates downstream analyses. To overcome this problem, we treated zeros as missing values and developed deep learning methods to impute the missing values, exploiting the dependence structure across genes and across cells. Specifically, our LATE (Learning with AuToEncoder) method trains an autoencoder directly on scRNA-seq data with random initial values of the parameters, whereas our TRANSLATE (TRANSfer learning with LATE) method further allows for the use of a reference gene expression data set to provide LATE with an initial set of parameter estimates. On both simulated and real data, LATE and TRANSLATE outperform existing scRNA-seq imputation methods. Importantly, LATE and TRANSLATE are highly scalable and can impute gene expression levels in over 1 million cells in just a few hours. We implemented our methods using Google TensorFlow and our software can run on a GPU as well as a CPU.

Download data

  • Downloaded 1,162 times
  • Download rankings, all-time:
    • Site-wide: 5,258 out of 59,758
    • In bioinformatics: 1,015 out of 6,035
  • Year to date:
    • Site-wide: 1,400 out of 59,758
  • Since beginning of last month:
    • Site-wide: 11,284 out of 59,758

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide

Sign up for the Rxivist weekly newsletter! (Click here for more details.)