Assessment of single cell RNA-seq normalization methods

UCSD researchers have assessed the performance of seven normalization methods for single cell RNA-seq using data generated from dilution of RNA samples. Their analyses showed that methods considering spike-in ERCC RNA molecules significantly outperformed those not considering ERCCs. This work provides a guidance of selecting normalization methods to remove technical noise in single cell RNA-seq data.


Hierarchical clustering of HBR and UHR RNA-seq samples with different normalization methods not considering ERCC. (a) DESeq with HBR samples; (b) RUVr with HBR samples (c-d) DESeq and RUVr with UHR samples.

Ding B, Zheng L, Wang W. (2016) Assessment of single cell RNA-seq normalization methods. bioRXiv [Epub ahead of print. [abstract]

One comment

  1. I am wondering if anyone knows how the cDNA synthesis was performed. The methods do not include any data, but the paper indicates a modified SMART protocol on the C1, an aRNA method, and the NuGen Ovation kit. The references for the SMART protocol does not mention the C1.

Leave a Reply

Your email address will not be published. Required fields are marked *


Time limit is exhausted. Please reload CAPTCHA.