Comparative performance of the BGISEQ-500 and Illumina HiSeq4000 sequencing platforms for transcriptome analysis in plants

Abstract

Background: The next-generation sequencing (NGS) technology has greatly facilitated genomic and transcriptomic studies, contributing significantly in expanding the current knowledge on genome and transcriptome. However, the continually evolving variety of sequencing platforms, protocols and analytical pipelines has led the research community to focus on cross-platform evaluation and standardization. As a NGS pioneer in China, the Beijing Genomics Institute (BGI) has announced its own NGS platform designated as BGISEQ-500, since 2016. The capability of this platform in large-scale DNA sequencing and small RNA analysis has been already evaluated. However, the comparative performance of BGISEQ-500 platform in transcriptome analysis remains yet to be elucidated. The Illumina series, a leading sequencing platform in China's sequencing market, would be a preferable reference to evaluate new platforms.

Methods: To this end, we describe a cross-platform comparative study between BGISEQ-500 and Illumina HiSeq4000 for analysis of Arabidopsis thaliana WT (Col 0) transcriptome. The key parameters in RNA sequencing and transcriptomic data processing were assessed in biological replicate experiments, using aforesaid platforms.

Results: The results from the two platforms BGISEQ-500 and Illumina HiSeq4000 shared high concordance in both inter- (correlation, 0.88-0.93) and intra-platform (correlation, 0.95-0.98) comparison for gene quantification, identification of differentially expressed genes and alternative splicing events. However, the two platforms yielded highly variable interpretation results for single nucleotide polymorphism and insertion-deletion analysis.

Conclusion: The present case study provides a comprehensive reference dataset to validate the capability of BGISEQ-500 enabling it to be established as a competitive and reliable platform in plant transcriptome analysis.

Keywords: Alternative splicing; BGISEQ-500; Differential expressed genes; Illumina HiSeq4000; Next-generation sequencing; Transcriptome.

Fig. 1

Schematic view of analytical pipeline of this study. SNP single nucleotide polymorphism, INDEL insertion–deletion, AS alternative splicing, DEG differentially expressed genes

Fig. 2

Comparison of sequencing quality among BGISEQ-500 PE75, BGISEQ-500 PE100 and HiSeq4000 PE100. a Base quality representation for clean reads and Q20. b Reads quality evaluation and mapping percentage. c Reads distribution along the relative position of genes

Fig. 3

Repeatability of gene detection and quantification among three sequencing approaches. a Venn diagram representation of gene detection. Expression density distribution (b), boxplot gene expression graph (c), high and low abundance transcripts quantification (d) for all the replicates tested by three sequencing approaches in this study

Fig. 4

Differentially expressed genes determination among three sequencing approaches. a Venn diagram representation of DEG calling in each sequencing approach. b Cross-platform comparison in DEG detection. c Pathway enrichment of each sequencing approach. Black, pathways enriched in all the three approaches; Red, pathways enriched in two approaches; Orange, pathways enriched in one approach. A, alpha-Linolenic acid metabolism; B, Anthocyanin biosynthesis; C, Biosynthesis of secondary metabolites; D, Biosynthesis of unsaturated fatty acids; E, Carotenoid biosynthesis; F, Cutin, suberine and wax biosynthesis; G, Flavonoid biosynthesis; H, Galactose metabolism; I, Glycerolipid metabolism; J, Indole alkaloid biosynthesis; K, Metabolic pathways; L, Other glycan degradation; M, Biosynthesis of secondary metabolites in phenylpropanoid pathway; N, Plant hormone signal transduction; O, Plant-pathogen interaction; P, Starch and sucrose metabolism; Q, Phenylpropanoid biosynthesis; R, Other terpenoid biosynthesis; S, Zeatin biosynthesis; T, MAPK signaling pathway; U, Peroxisome; V, Fatty acid metabolism; W, Pentose and glucuronate interconversions; X, Phenylalanine metabolism

Fig. 5

Comparison of alternative spliced events identification. Venn diagrams representation of AS events identification in DMSO- (a) and ABA-treated (b) samples by each sequencing approach. Venn diagrams to represent c AS events in DMSO-treated and d ABA-treated samples. ATS alternative transcription start, APA alternative polyadenylation, AE5′ alternative 5′ splice site, AE3′ alternative 3′ splice site