Comparing the normalization methods for the differential analysis of Illumina high-throughput RNA-Seq data

Abstract

BackgroundRecently, rapid improvements in technology and decrease in sequencing costs have made RNA-Seq a widely used technique to quantify gene expression levels. Various normalization approaches have been proposed, owing to the importance of normalization in the analysis of RNA-Seq data. A comparison of recently proposed normalization methods is required to generate suitable guidelines for the selection of the most appropriate approach for future experiments.ResultsIn this paper, we compared eight non-abundance (RC, UQ, Med, TMM, DESeq, Q, RPKM, and ERPKM) and two abundance estimation normalization methods (RSEM and Sailfish). The experiments were based on real Illumina high-throughput RNA-Seq of 35- and 76-nucleotide sequences produced in the MAQC project and simulation reads. Reads were mapped with human genome obtained from UCSC Genome Browser Database. For precise evaluation, we investigated Spearman correlation between the normalization results from RNA-Seq and MAQC qRT-PCR values for 996 genes. Based on this work, we showed that out of the eight non-abundance estimation normalization methods, RC, UQ, Med, TMM, DESeq, and Q gave similar normalization results for all data sets. For RNA-Seq of a 35-nucleotide sequence, RPKM showed the highest correlation results, but for RNA-Seq of a 76-nucleotide sequence, least correlation was observed than the other methods. ERPKM did not improve results than RPKM. Between two abundance estimation normalization methods, for RNA-Seq of a 35-nucleotide sequence, higher correlation was obtained with Sailfish than that with RSEM, which was better than without using abundance estimation methods. However, for RNA-Seq of a 76-nucleotide sequence, the results achieved by RSEM were similar to without applying abundance estimation methods, and were much better than with Sailfish. Furthermore, we found that adding a poly-A tail increased alignment numbers, but did not improve normalization results.ConclusionSpearman correlation analysis revealed that RC, UQ, Med, TMM, DESeq, and Q did not noticeably improve gene expression normalization, regardless of read length. Other normalization methods were more efficient when alignment accuracy was low; Sailfish with RPKM gave the best normalization results. When alignment accuracy was high, RC was sufficient for gene expression calculation. And we suggest ignoring poly-A tail during differential gene expression analysis.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0778-7) contains supplementary material, which is available to authorized users.

Human iPSC Glial Mouse Chimeras Reveal Glial Contributions to Schizophrenia

“…G, Heat map representation of 4 conserved differentially expressed genes that are associated to module 4 ( light blue in B, 10.2%), with annotations related to myelination and lipid biosynthesis. The absolute expression in heat maps is shown in UQ-normalized, log2-transformed counts (Li et al, 2015). …”

Construction and Optimization of a Large Gene Coexpression Network in Maize Using RNA-Seq Data

“…Expression data from distinct sources and experiments can be highly variable because of hybridization artifacts in microarray or variable sequencing depth in RNA-Seq. Many methods have been successfully used for normalizing both microarray and RNA-Seq data to correct for potential biases (Lim et al, 2007;Dillies et al, 2013;Li et al, 2015b). To find an optimal normalization method for building a maize GCN from RNA-Seq data, three widely used normalization methods were compared.…”

“…Therefore, genomewide expression results generated by either technique need to be normalized before analysis (Dillies et al, 2013;Li et al, 2015b). Variance stabilizing transformation (VST), counts per million (CPM), and reads per kilobase million (RPKM) are three popular normalization methods for RNA-Seq experiments (Mortazavi et al, 2008;Anders and Huber, 2010;Rau et al, 2013).…”

Androgen receptor-deficient islet β-cells exhibit alteration in genetic markers of insulin secretion and inflammation. A transcriptome analysis in the male mouse

“…If we use kilobase as the unit for L and million reads as the unit for N , this estimation is called reads per kilobase of transcript per million mapped reads (RPKM), which is the most widely used RNA-seq normalization method (Li, Piao, Shon, & Ryu, 2015). The individual transcript files generated by Cufflinks for each sample were merged into a single gene annotation file, which was then used to perform a DE analysis with the Cufflinks routine, Cuffdiff.…”