Introgressions lead to reference bias in wheat RNA-seq analysis

Area of Science:

• Genomics
• Bioinformatics
• Plant Science

Background:

• RNA sequencing (RNA-seq) is crucial for gene expression analysis.
• Reference bias, where non-reference alleles are poorly quantified, impacts RNA-seq accuracy.
• This bias is unexplored in complex polyploid genomes, such as wheat, which contain introgressions from wild relatives.

Purpose of the Study:

• To investigate reference bias in hexaploid wheat RNA-seq data.
• To identify the role of introgressed genes in causing reference bias.
• To develop a method to mitigate reference bias in polyploid transcriptomics.

Main Methods:

• Utilized simulated and experimental RNA-seq data from hexaploid wheat.
• Developed a pantranscriptome reference incorporating gene models from ten wheat genome assemblies.
• Applied the novel method to reduce reference bias in RNA-seq alignment.

Main Results:

• Confirmed widespread reference bias in wheat RNA-seq alignment, primarily due to divergent introgressed genes.
• Demonstrated that reference bias leads to underestimation of gene expression and inaccurate homoeologue expression balance.
• The developed pantranscriptome approach effectively reduced reference bias.

Conclusions:

• Introgressions in polyploid genomes like wheat significantly contribute to reference bias in RNA-seq.
• Researchers should exercise caution when using non-sample reference genomes for RNA-seq alignment.
• The presented pantranscriptome method offers a scalable solution to improve RNA-seq accuracy in complex genomes.