Abstract
BackgroundThe number of panicles per plant, number of grains per panicle, and 1000-grain weight are important factors contributing to the grain yield per plant in rice. The Rice Diversity Panel 1 (RDP1) contains a total of 421 purified, homozygous rice accessions representing diverse genetic variations within O. sativa. The release of High-Density Rice Array (HDRA, 700 k SNPs) dataset provides a new opportunity to discover the genetic variants of panicle architectures in rice.ResultsIn this report, a new method genome-phenome wide association study (GPWAS) was performed with 391 individuals and 27 traits derived from RDP1 to scan the relationship between the genes and multi-traits. A total of 1985 gene models were linked to phenomic variation with a p-value cutoff of 4.49E-18. Besides, 406 accessions derived from RDP1 with 411,066 SNPs were used to identify QTLs associated with the total spikelets number per panicle (TSNP), grain number per panicle (GNP), empty grain number per panicle (EGNP), primary branch number (PBN), panicle length (PL), and panicle number per plant (PN) by GLM, MLM, FarmCPU, and BLINK models for genome-wide association study (GWAS) analyses. A total of 18, 21, 18, 17, 15, and 17 QTLs were identified tightly linked with TSNP, GNP, EGNP, PBN, PL, and PN, respectively. Then, a total of 23 candidate genes were mapped simultaneously using both GWAS and GPWAS methods, composed of 6, 4, 5, 4, and 4 for TSNP, GNP, EGNP, PBN, and PL. Notably, one overlapped gene (Os01g0140100) were further investigated based on the haplotype and gene expression profile, indicating this gene might regulate the TSNP or panicle architecture in rice.ConclusionsNearly 30 % (30/106) QTLs co-located with the previous published genes or QTLs, indicating the power of GWAS. Besides, GPWAS is a new method to discover the relationship between genes and traits, especially the pleiotropy genes. Through comparing the results from GWAS and GPWAS, we identified 23 candidate genes related to panicle architectures in rice. This comprehensive study provides new insights into the genetic basis controlling panicle architectures in rice, which lays a foundation in rice improvement.
Highlights
The number of panicles per plant, number of grains per panicle, and 1000-grain weight are important factors contributing to the grain yield per plant in rice
Through comparing the results from Genome-Wide Association Study (GWAS) and Genome-Phenome Wide Association Study (GPWAS), we identified 23 candidate genes related to panicle architectures in rice
The results exhibited that the population could be divided into five groups, which was corresponding to the Principal component analysis (PCA) analysis
Summary
The number of panicles per plant, number of grains per panicle, and 1000-grain weight are important factors contributing to the grain yield per plant in rice. Yield increasing is an important goal in the rice improvement process, which could be divided into three major components - number of panicles per plant, number of grains per panicle, and 1000-grain weight [2]. Several genes have been reported regulating the grains per panicle in rice, including Gn1a [4], GNP1 [5], GAD1 [6], An-1 [7], OsCBL8 [8], OsDim1 [8], OsMADS18 [9], PAY1 [10], and SAPK2 [11]. When the GNP1 was upregulated, the activity of cytokinin was increased because of a KNOX-mediated transcriptional feedback loop, resulting in an increased grain number and grain yield in rice [5]. GS3 is a major gene regulating the grain size in rice grain [16], which has been used to improve yield with the CRISPR-Cas system [17]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
