Meta-GWAS analyses
Meta-analyses of GWAS were performed based on four datasets’ (two locations for each population) GWAS results (Additional file 1: Figures S1-S4). Manhattan plots and quantile-quantile plots of meta-GWAS are shown in Fig. 1. A total of 3592 significant loci were identified (Additional file 2: Table S1), including 2450, 1116, 23 and, 3 significant associated loci were separately detected for GPP, KGW, TP, and YD, which were distributed on all of the rice chromosomes except for chromosome 10. According to the information of RAP-DB (http://rapdb.dna.affrc.go.jp/), candidate genes were searched in a genomic region of 200KB around the associated SNPs (Additional file 1: Table S2). We discovered 7, 7 and 3 cloned genes separately associated with GPP, KGW and TP. A total of three candidate genes associated with different traits, among which OsBZR1 (Zhu et al. 2015) and OsSPL14 (Jiao et al. 2010) have been reported previously, and Os02g0106966 was novelty discovered. Both OsBZR1 and OsSPL14 were detected in KGW and GPP, Os02g0106966 was detected in KGW and TP. In this study, only 3 significant loci for YD were detected, but 3589 significant loci for the component traits were detected. It may be because rice yield has a low heritability which mainly affected by many minor-effect genes, the low heritability of rice yield is also showed in our previous study (Xu et al. 2018). These results suggested that selecting the component traits of yield as a complementary route to improve rice production is recommended.
The genetic relationship between GPP and YD
As required for MR analysis, a total of 2450 SNPs reached genome-wide significance for GPP (P<1E-06) in meta-analyses of GWAS, among which six SNPs were selected as instrumental variables to estimate the genetic relationship between GPP and YD (Table 1). The six SNPs were not associated with KGW or TP (P>0.05), and no evidence of LD between them (all pairwise r2<0.01). In MR analysis, these loci mainly had a positive genetic effect on yield through GPP and a positive genetic relationship between GPP and YD were observed with the IVW method (Table 4, Fig. 2a). One standard deviation (SD) genetic higher GPP was associated with a 0.086 SD higher YD (Beta=0.086, 95% CI: 0.030~0.141, P=0.003). In sensitivity analyses, the Cochran's Q-test illustrated no obvious heterogeneity (I2=5%, P=0.38). The weighted median method also showed GPP had a positive effect on YD (Beta=0.081, 95% CI: 0.009~0.152, P=0.028). MR-Egger regression indicated no evidence of directional pleiotropy for the associations of GPP with YD (intercept=1.387,P=0.061). It is worth noting that some cloned genes were detected in the meta-GWAS on GPP, the phenotype of transgenic plants with these genes had a similar phenomenon. For example, the OsSPL14 mutant produced more grain number per panicle, enhanced rice yield (Jiao et al. 2010). Compared with the control non-transgenic plants, the over-expression of OsBZR1 plants showed the 1000-grain weight was increased by about 3.4% and the spikelet number per panicle was increased 21.9%, that resulting in enhanced yield (Zhu et al. 2015). The cd1 mutant exhibited a variety of phenotypic traits, such as a reduction in grain number and panicle length, the biomass was lower than that of the wild type (Luan et al. 2011).
Table 1 Information about instrumental variables
SNP
|
Chromosome
|
Position
|
GPP
|
|
YD
|
Beta
|
P-value
|
|
Beta
|
P-value
|
chr03_29979498
|
3
|
29979498
|
-18.724
|
1.12E-07
|
|
-0.633
|
0.543
|
chr03_898774
|
3
|
898774
|
18.953
|
2.92E-07
|
|
1.254
|
0.299
|
chr05_7226049
|
5
|
7226049
|
-7.242
|
2.25E-08
|
|
-1.329
|
0.005
|
chr08_25257522
|
8
|
25257522
|
-16.559
|
8.58E-07
|
|
1.178
|
0.553
|
chr09_12464309
|
9
|
12464309
|
-7.538
|
4.49E-07
|
|
-0.974
|
0.066
|
chr12_22633431
|
12
|
22633431
|
15.738
|
3.01E-07
|
|
1.439
|
0.158
|
Note: All the SNP markers are named after the chromosome _ position
The genetic relationship between KGW and YD
As required for MR analysis, a total of 1116 SNPs reached genome-wide significance for KGW (P<1E-06) in meta-analyses of GWAS, among which eleven SNPs were selected as instrumental variables to estimate the genetic relationship between KGW and YD (Table 3). These SNPs were not associated with GPP or TP (P>0.05), and no evidence of LD between them (all pairwise r2<0.01). In MR analysis, we observed that a part of SNP for KGW had a positive effect on YD, a part of SNP for KGW had a negative effect on YD (Table 2, Fig. 2b). To further understand to genetic relationship between KGW and YD, the SNPs with different direction of genetic effects are studied separately in our study. These loci with positive effect showed that KGW had a positive effect on yield, while these loci with negative effect showed that KGW had no significant negative effect on yield (Fig. 3). In sensitivity analyses, the Cochran's Q-test illustrated no obvious heterogeneity (I2=0%). The weighted median method also confirmed the results of the IVW method. MR-Egger regression indicated no evidence of pleiotropy for the associations of KGW with YD (Table 4). The cloned gene GW2 was detected in the meta-GWAS on KGW has been reported have the potential to enhance rice yield (Song et al. 2007).
Table 2 Information about instrumental variables
SNP
|
Chromosome
|
Position
|
KGW
|
|
YD
|
Beta
|
P-value
|
|
Beta
|
P-value
|
chr01_3547491
|
1
|
3547491
|
0.706
|
7.14E-08
|
|
-0.036
|
0.948
|
chr01_5524333
|
1
|
5524333
|
1.018
|
6.56E-07
|
|
0.310
|
0.840
|
chr02_1118809
|
2
|
1118809
|
0.786
|
1.56E-09
|
|
1.030
|
0.082
|
chr02_334316
|
2
|
334316
|
-1.474
|
2.38E-07
|
|
0.948
|
0.505
|
chr02_7792121
|
2
|
7792121
|
-1.256
|
9.21E-08
|
|
0.029
|
0.980
|
chr03_17810847
|
3
|
17810847
|
0.651
|
4.74E-07
|
|
0.711
|
0.394
|
chr03_33060865
|
3
|
33060865
|
-0.505
|
2.03E-08
|
|
-0.978
|
0.027
|
chr04_13785932
|
4
|
13785932
|
0.739
|
3.16E-07
|
|
0.623
|
0.406
|
chr05_16393143
|
5
|
16393143
|
0.466
|
4.71E-07
|
|
0.029
|
0.955
|
chr07_23215227
|
7
|
23215227
|
0.952
|
1.09E-07
|
|
-0.308
|
0.715
|
chr08_25464238
|
8
|
25464238
|
-0.875
|
2.01E-09
|
|
-0.004
|
0.997
|
The genetic relationship between TP and YD
As required for MR analysis, a total of 23 SNPs reached genome-wide significance for TP (P<1E-06) in meta-analyses of GWAS, among which three SNPs were selected as instrumental variables to estimate the causal relationship between TP and YD (Table 3). These SNPs were not associated with KGW or GPP (P>0.05), and no evidence of LD between them (all pairwise r2<0.01). In MR analysis, these loci had a positive genetic effect on yield through TP and a positive causal relationship between TP and YD were observed with the IVW method (Table 4, Fig. 2c), 1 SD genetic higher TP was associated with a 1.865 SD higher YD (Beta=1.865, 95% CI: 1.035~2.694, P<0.0001). Compared with KGW (Beta=1.016) and GPP (Beta=0.086), TP (Beta=1.865) has a greater effect on yield. In sensitivity analyses, Cochran's Q-test illustrated no obvious heterogeneity (I2=0%, P=0.43). The weighted median method also showed TP had a positive effect on YD (Beta=1.54, 95% CI: 0.353~2.727, P=0.011). MR-Egger regression indicated no evidence of directional pleiotropy for the associations of TP with YD (intercept=0.046, P=0.968). The cloned gene OsPIN2 was detected in the meta-GWAS on TP. Chen et al. (2012) found that the OsPIN2 transgenic plants had more effective tiller number, lower 1000-grain weight, and higher yield.
Table 3 Information about instrumental variables
SNP
|
Chromosome
|
Position
|
TP
|
|
YD
|
Beta
|
P-value
|
|
Beta
|
P-value
|
chr02_21604477
|
2
|
21604477
|
-1.279
|
8.25E-07
|
|
-1.977
|
0.043
|
chr06_1578700
|
6
|
1578700
|
-0.639
|
4.52E-07
|
|
-1.691
|
3.21E-04
|
chr11_26492375
|
11
|
26492375
|
-0.507
|
7.04E-08
|
|
-0.720
|
0.043
|
Table 4 MR results of the relationship between yield and its component traits
Trait
|
Methods
|
Beta
|
95% CI
|
P
|
GPP
|
IVW
|
0.086
|
0.030~ 0.141
|
0.003
|
Weighted median
|
0.081
|
0.009 ~ 0.152
|
0.028
|
MR-Egger
|
-0.029
|
-0.160~0.103
|
0.668
|
MR-Egger(intercept)
|
1.387
|
-0.063~2.836
|
0.061
|
KGW (positive)
|
IVW
|
1.016
|
0.242~1.791
|
0.010
|
Weighted median
|
1.123
|
0.122~2.124
|
0.028
|
MR-Egger
|
0.480
|
-2.743~3.704
|
0.770
|
MR-Egger(intercept)
|
0.349
|
-1.690~2.388
|
0.737
|
KGW
(negative)
|
IVW
|
-0.233
|
-1.092~0.626
|
0.595
|
Weighted median
|
-0.156
|
-1.150~0.839
|
0.759
|
MR-Egger
|
-0.710
|
-3.853~2.434
|
0.658
|
MR-Egger(intercept)
|
0.464
|
-2.480~3.407
|
0.757
|
TP
|
IVW
|
1.865
|
1.035~2.694
|
<0.0001
|
Weighted median
|
1.540
|
0.353~2.727
|
0.011
|
MR-Egger
|
1.797
|
-1.633~5.228
|
0.304
|
MR-Egger(intercept)
|
0.046
|
-2.165~2.256
|
0.968
|
Note: CI confidence intervals, P statistically significant associations with a P<0.05.
Loci for component traits had an indirect effect on yield
We identified five significant loci that had an indirect effect on yield by MR analyses (Fig. 2, Additional file 1: Table S3). Among them, the SNP chr05_7226049 (Fig. 2a) for GPP had an indirect effect on yield and located nearby the cloned gene OsPYL11. Kim et al. (2014) reported that compared with the control plants, the transgenic plants overexpressing OsPYL11 showed no significant difference in tiller number, but the yield was severely reduced. Our study indicated the yield severely reduced may be caused by the number of grains decreased. The SNP chr03_33060865 (Fig. 2b) for KGW is in the vicinity of the cloned gene EL1, which is a key regulator of the gibberellin response, Kwon et al. (2015) discovered the plants that loss of EL1 showed the 500-grain weight and yield significantly reduced. The SNP chr06_1578700 (Fig. 2c) for TP closed to the D62 (a gene regulating tillers). Li et al. (2010) found that the tiller number of D62 mutant rice was less than that of the wild type. The SNPs chr02_21604477 and chr11_26492375 for TP also had indirect effects on yield (Fig. 2c, Table S3), which were first detected in our research. These findings provided new information for further improve rice yield potential.
Pyramiding superior alleles of significant loci
The average yield performance of F1 lines with different superior allele numbers of significant loci with direct effect, indirect effect, and direct plus indirect effect were shown in Fig. 4. Three loci had a direct effect on yield were detected in the meta-GWAS on YD (Additional file 1: Table S3), among them, the average yield of the lines without superior alleles was 41.29g, and the average yield of the lines with one superior allele was 44.26g (Fig. 4a, Additional file 1: Table S4). The superior alleles of five loci that had an indirect effect on yield were also pyramided in the study. The results showed that the average yield of F1 lines with 0 to 4 superior alleles was: 42.22g, 42.75g, 42.76g, 44.54g, 47.49g, respectively. In general, the yield of F1 hybrids rises with increases of the superior alleles (Fig. 4b, Additional file 1: Table S4). A similar phenomenon also found in pyramiding the direct plus indirect loci (Fig. 4c, Additional file 1: Table S4). Other research reported that the phenotype performance improved by pyramiding the superior alleles of loci associated with agronomic traits (Huang et al. 2015), our results suggested the yield enhanced also by pyramiding the superior alleles of loci that had an indirect effect on yield. Hybrid lines pyramiding all the superior alleles of direct (3 loci) and indirect loci (5 loci) not be observed in this study. Our results indicated rice production improved with increased of the superior alleles, it is possible that a combination of direct and indirect effects will better contribute to the yield potential of rice.