Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Nov 15;29(22):2943-5.
doi: 10.1093/bioinformatics/btt511. Epub 2013 Aug 31.

SAMstrt: statistical test for differential expression in single-cell transcriptome with spike-in normalization

Shintaro Katayama  1 , Virpi Töhönen, Sten Linnarsson, Juha Kere
Affiliations

SAMstrt: statistical test for differential expression in single-cell transcriptome with spike-in normalization

Shintaro Katayama et al. Bioinformatics. .

Abstract

Motivation: Recent transcriptome studies have revealed that total transcript numbers vary by cell type and condition; therefore, the statistical assumptions for single-cell transcriptome studies must be revisited. SAMstrt is an extension code for SAMseq, which is a statistical method for differential expression, to enable spike-in normalization and statistical testing based on the estimated absolute number of transcripts per cell for single-cell RNA-seq methods.

Availability and implementation: SAMstrt is implemented on R and available in github (https://github.com/shka/R-SAMstrt).

Contact: shintaro.katayama@ki.se

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.
(A) Aligned reads and the detected features in 45 mES, 44 MEF and 24 of 50 pg human brain total RNAs (50 pg) by STRT. The features are known genes, repeat elements and spike-in molecules; there were 25 286 features for mouse and 25 665 features for human. (B) False discovery features in 100 trials between SAMseq and SAMstrt. In each trial, 24 of 50 pg human brain total RNAs were separated randomly into two groups, and then each method compared the two groups. Although number of differentially expressed features should be zero in all trials if no technical variations, the false discovery features showed statistical difference (FDR < 1%). (C) Comparison of transcripts per 50 pg human brain total RNAs per feature, between the former 12 samples and the latter 12 samples. This is one representative comparison in the trials for panel B, and there were no significantly different features by the proposed SAMstrt (FDR < 0.01%). Gray color gradation denotes density of features in the scatter plot, and expression level of each feature is represented by median. Dashed diagonal line denotes equivalent expression between two samples. (D) Comparisons of normalized expression levels per cell per feature, between 45 mES and 44 MEF cells, by SAMseq. Usage of the gray color gradation and the dashed diagonal line are same with the panel C. Points are differentially expressed features (FDR < 0.01%). (E) Sum of the normalized expression values of all features by samples and the comparison by the methods. (F) Comparison of transcripts per cell per feature, between 45 mES and 44 MEF cells, by SAMstrt. Usage of the gray color gradation, the dashed diagonal line, and the points are same with the panel D. SAMseq added uniform random numbers between 0 and 0.1 to all values to avoid ties (Li and Tibshirani, 2011), therefore, features which are the most bottom expression level in the panel C, D and F are no expression, or less than detection limit. The normalized expression values by SAMstrt at the panel C and F are moreover estimated transcripts per cell based on the initial concentration of the spike-in molecules
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Anders S, Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010;11:R106. - PMC - PubMed
    1. Bullard JH, et al. Evaluation of statistical methods for normalization and differential expression in mRNA-Seq experiments. BMC Bioinformatics. 2010;11:94. - PMC - PubMed
    1. Chubb JR, et al. Transcriptional pulsing of a developmental gene. Curr. Biol. 2006;16:1018–1025. - PMC - PubMed
    1. Dobson AT, et al. The unique transcriptome through day 3 of human preimplantation development. Hum. Mol. Genet. 2004;13:1461–1470. - PubMed
    1. Hashimshony T, et al. CEL-Seq: single-cell RNA-seq by multiplexed linear amplification. Cell Rep. 2012;2:666–673. - PubMed

Publication types

MeSH terms

-