Multiomic insights into sucrose accumulation in sugarcane

Abstract

Sugarcane holds significant economic importance in sugar and biofuel production. Despite extensive research, understanding highly quantitative traits remains challenging due to its complex genomic landscape. We conducted a multiomic investigation to elucidate the genetic architecture and molecular mechanisms governing sugarcane sucrose accumulation. Using a biparental cross and a genetically diverse collection of sugarcane genotypes, we evaluated the soluble solids (Brix) and sucrose content (POL) across various years. Both populations were genotyped using a genotyping-by-sequencing approach. Genotype‒phenotype associations were established using a combination of traditional linear mixed-effect models and machine learning algorithms. Furthermore, we conducted an RNA sequencing experiment on genotypes exhibiting distinct Brix and POL profiles across different developmental stages. Differentially expressed genes (DEGs) potentially associated with variations in sucrose accumulation were identified. All findings were integrated through gene coexpression network analyses. Strong correlations among the evaluated characteristics were observed, with estimates of modest to high heritabilities. By leveraging a broad set of single-nucleotide polymorphisms (SNPs) identified for both populations, we identified several SNPs potentially linked to phenotypic variance. Our examination of genes close to these markers facilitated the association of such SNPs with DEGs for contrasting sucrose levels. Through the integration of these results with a gene coexpression network, we delineated a set of genes potentially involved in the regulatory mechanisms of sucrose accumulation. Our findings constitute a significant resource for biotechnology and plant breeding initiatives. Furthermore, our genotype‒phenotype association models hold promise for application in genomic selection, offering valuable insights into the molecular underpinnings governing sucrose accumulation in sugarcane.

Keywords: Saccharum spp.; GWAS; Gene coexpression networks; Genotyping-by-sequencing; Machine learning; RNA-Seq.