Transcriptome brings variations of gene expression, alternative splicing, and structural variations into gene-scale trait dissection in soybean

Abstract

Genome-wide association study (GWAS) identifies trait-associated loci, but due in part to slow decay of linkage disequilibrium (LD), identifying the causal genes can be a bottleneck. Transcriptome-wide association study (TWAS) addresses this by identifying gene expression-phenotype associations or integrating gene expression quantitative trait loci (eQTLs) with GWAS results. Here, we used self-pollinated soybean as a model to evaluate the application of TWAS in the genetic dissection of traits in plant species that exhibit slow LD decay. The first RNA-Seq analysis of a soybean diversity panel was conducted, which identified the genetic regulation of 29,286 genes. Different TWAS solutions were less affected by LD and robust with source of expression that identified known genes related to traits from different development stages and tissues. A novel gene named pod color L2 was identified via TWAS and functionally validated by genome editing. Our introduction of the new exon proportion feature significantly improves the capture of expression variations resulting from structural variations and alternative splicing. As a result, the genes identified by our TWAS approach exhibited a diverse range of causal variations, including SNP, insertion/deletion, gene fusion, copy number variation, and alternative splicing. Using our TWAS approach, we identified genes associated with flowering time, including both previously known candidates and novel genes that had not been linked to this trait before, providing complementary insights with GWAS. In summary, this study supports the application of TWAS for candidate gene identification in species with low rates of LD decay.

Population‐level gene expression can repeatedly link genes to functions in maize

“…Another frequently voiced concern regarding the reuse of transcriptome-wide association data is whether gene expression must be measured from the tissue where the phenotype of interest is apparently. Fortunately, several recent studies have demonstrated that analysis using gene expression from off-target tissues can still accurately identify known true-positive causal genes (Li et al, 2021(Li et al, , 2023Sun et al, 2023). Consistent with these earlier reports, two of the genes whose expression in mature leaf tissue was most closely associated with flowering time-zmm4 and zmm15-are genes known to act in the meristem rather than mature leaf tissue (Danilevskaya, Meng, Selinger, et al, 2008).…”

“…TWAS identifies complementary rather than redundant sets of genes to those identified via GWAS for the same phenotypes in the same populations. The expression level of an individual gene can integrate the signals from multiple upstream regulatory variants, each too small or too rare to be linked directly to variation in the phenotype of interest (Li et al, 2023). TWAS based on direct measurements of gene expression typically identifies specific candidate genes rather than intervals containing multiple genes, even in species or populations with slow decay of linkage disequilibrium across the genome (Li et al, 2023).…”

“…This raises concerns about how much, if any, potential exists to reuse transcriptome-wide expression datasets to identify genes linked to variation in different traits in different environments. Multiple studies have demonstrated that gene expression from non-target tissues and non-target environments can identify true-positive causal genes (Hirsch et al, 2014;Li et al, 2021Li et al, , 2023. However, the reuse of transcript abundance data across multiple studies conducted using the same population remains less widely adopted than the reuse of genetic marker data across studies.…”

“…The expression level of an individual gene can integrate the signals from multiple upstream regulatory variants, each too small or too rare to be linked directly to variation in the phenotype of interest (Li et al, 2023). TWAS based on direct measurements of gene expression typically identifies specific candidate genes rather than intervals containing multiple genes, even in species or populations with slow decay of linkage disequilibrium across the genome (Li et al, 2023). However, the terminology TWAS has also been applied to methodologies that use genetic marker data to impute gene expression and seek to link that imputed gene expression data to phenotype.…”

PH13 improves soybean shade traits and enhances yield for high-density planting at high latitudes

“…The raw sequence data 24 and gene expression data 63 reported elsewhere are available at NCBI Sequence Read Archive under accession PRJNA681974 and the Genome Sequence Archive (GSA) database of the BIG Data Center under accession PRJCA014188, respectively. The sequences of the three PH13 alleles reported in this study are available at GenBank of NCBI: PH13 H1 /OR637868, PH13 H2 /OR637869, and PH13 H3 / OR637870.…”