关键词:棉花( Gossypium hirsutum L.); 区域试验; 遗传力; 噪信比; 重复次数; 试点数量; 优化配置 Design of Test Location Number and Replicate Frequency in the Regional Cotton Variety Trials in China XU Nai-Yin1, JIN Shi-Qiao2, LI Jian1 1 Institute of Industrial Crops, Jiangsu Academy of Agricultural Sciences / Key Laboratory of Cotton and Rapeseed, Ministry of Agriculture, Nanjing 210014, China
2 National Extension and Service Center of Agricultural Technology, Beijing 100125, China
Fund:This study was supported by the Major Project of China on New Varieties of GMO Cultivation (2012ZX08013015) and the Project from the National Extension and Service Center of Agricultural Technology (012022911108) AbstractThe test location number and the replicate frequency in regional crop trials are important factors in determining both the trial heritability and cultivar selection efficiency. The test location number and replicate frequency for three national cotton regional trials in China were studied using experimental data during the last 15 years according to changes of trial heritability with the increase of test locations and replicates within trials in 2000-2014. The results indicated that three replicates are sufficient to achieve 0.75 of within-trial heritability. The current test locations in the Yangtze River Valley, the Yellow River Valley and the Northwest Inland regions are sufficient to achieve 0.75 of cross-trial heritability. Considering the importance of the regional trials in recommending cotton varieties and the possible trial cancellation due to poor field managements, natural disasters or other non-artificial factors, the optimum number of test locations proposed for the Yangtze River Valley should be maintained at the current level of 20 locations with H = 0.90 to ensure enough credibility of regional trials, while that proposed for the Yellow River Valley and the Northwest Inland cotton regions should be increased to 27 and 19 locations with heritability level of 0.90 and 0.85, respectively. The conclusion will provide a theoretical guidance for the optimal configuration of national cotton regional trials and also act as a reference for the rational layout of regional trials in other crops.
Keyword:Cotton ( Gossypium hirsutum L.); Regional trial; Heritability; Noise-signal quotient; Frequency of replicates; Number of test locations; Optimal allocation Show Figures Show Figures
表1 2000-2014年我国棉花品种区域试验组数、品种数和试点数统计表 Table 1 Summary statistics of trial group, number of cultivars and locations used in regional cotton variety trials from 2000 to 2014
年份 Year
黄河流域棉区 Yellow River Valley
西北内陆棉区 Northwest Inland
长江流域棉区 Yangtze River Valley
全国 Nationwide
组数 Group
品种 Cultivar
试点 Site
组数 Group
品种 Cultivar
试点 Site
组数 Group
品种 Cultivar
试点 Site
组数 Group
品种 Cultivar
试点 Site
2000
1
8
19
2
26
11
1
10
16
4
44
46
2001
1
9
13
2
24
12
1
10
19
4
43
44
2002
1
11
11
2
19
14
1
10
19
4
40
44
2003
1
9
12
2
19
15
2
17
30
5
45
57
2004
2
22
36
2
19
13
2
18
35
6
59
84
2005
3
32
51
2
18
14
3
30
52
8
80
117
2006
4
46
70
2
18
13
3
31
51
9
95
134
2007
4
48
64
2
15
15
3
31
51
9
94
130
2008
4
37
59
2
13
14
3
34
45
9
84
118
2009
4
37
64
2
18
15
4
42
70
10
97
149
2010
4
36
54
2
18
12
4
40
70
10
94
136
2011
4
38
76
2
13
13
4
36
72
10
87
161
2012
3
30
60
2
24
13
4
38
72
9
92
145
2013
4
37
79
2
25
15
4
32
75
10
94
169
2014
4
32
79
2
27
18
3
24
57
9
83
154
平均Mean
2.9
28.8
49.8
2.0
19.7
13.8
2.8
26.9
48.9
7.7
75.4
112.5
合计Total
44
432
747
30
296
207
42
403
734
116
1131
1688
Trials groups include the mid-maturing conventional, the mid-maturing hybrid in the Yellow River Valley, the mid-maturing in the Yangtze River Valley, and the early-maturing and the early-medium maturing in the Northwest Inland regions. 表中列出了2000-2014年期间黄河流域棉区中熟常规棉和中熟杂交棉组、长江流域棉区中熟棉组和西北内陆棉区的早熟组与早中熟组国家棉花区域试验。
表1 2000-2014年我国棉花品种区域试验组数、品种数和试点数统计表 Table 1 Summary statistics of trial group, number of cultivars and locations used in regional cotton variety trials from 2000 to 2014
表2 2011-2014年全国棉花单年单点区试在遗传力为0.75时所需要重复数的次数分布 Table 2 Frequency distribution of replicate needed to achieve 0.75 of within-trial heritability in the national cotton regional trials from 2011 to 2014
需要重复数 Range of Nr needed
长江流域 Yangtze River Valley
黄河流域 Yellow River Valley
西北内陆 Northwest Inland
全国 Nationwide
次数 N
频数 Freq.(%)
平均值 Mean
次数 N
频数 Freq.(%)
平均值 Mean
次数 N
频数 Freq.(%)
平均值 Mean
次数 N
频数 Freq.(%)
平均值 Mean
Nr ≤ 1
152
55.07
1.00
178
60.54
1.00
34
57.63
1.00
364
57.87
1.00
1 < Nr ≤ 2
59
21.38
1.50
58
19.73
1.49
12
20.34
1.44
129
20.51
1.49
2 < Nr ≤ 3
18
6.52
2.48
30
10.20
2.50
6
10.17
2.53
54
8.59
2.50
3 < Nr≤ 4
12
4.35
3.50
7
2.38
3.67
1
1.69
3.40
20
3.18
3.56
4 < Nr≤ 5
7
2.54
4.60
4
1.36
4.38
2
3.39
4.50
13
2.07
4.52
5 < Nr≤ 6
8
2.90
5.55
5
1.70
5.40
1
1.69
5.20
14
2.23
5.47
6 < Nr≤ 9
9
3.26
7.04
2
0.68
7.55
0
0
—
11
1.75
7.14
Nr > 9
11
3.99
13.85
10
3.40
17.06
3
5.08
19.73
24
3.82
15.92
总计Total
276
100.00
2.24
294
100.00
2.02
59
100.00
2.43
629
100.00
2.16
Range of Nr means the replicates needed to achieve 0.75 of heritability in half-open intervals, Nr stands for replicate frequency. 重复数范围是指在H=0.75时所需要的试验重复次数的分布范围, 用Nr表示。
表2 2011-2014年全国棉花单年单点区试在遗传力为0.75时所需要重复数的次数分布 Table 2 Frequency distribution of replicate needed to achieve 0.75 of within-trial heritability in the national cotton regional trials from 2011 to 2014
表3 2000-2014年全国棉花单年多点区试在遗传力为0.75时所需要试点数量的次数分布 Table 3 Frequency distribution of test location number needed at 0.75 of cross-trial heritability in the cotton regional trials from 2000 to 2014
需要试点数 Range of Ne needed
长江流域 Yangtze River Valley
黄河流域 Yellow River Valley
西北内陆 Northwest Inland
全国 Nationwide
次数 N
频数 Freq.(%)
平均值 Mean
次数 N
频数 Freq.(%)
平均值 Mean
次数 N
频数 Freq.(%)
平均值 Mean
次数 N
频数 Freq.(%)
平均值 Mean
Ne ≤ 5
10
23.81
3.70
13
29.55
3.65
5
16.67
3.50
28
24.14
3.64
5 < Ne ≤ 10
29
69.05
6.98
15
34.09
7.06
12
40.00
7.05
56
48.28
7.02
10 < Ne ≤ 15
2
4.76
11.05
7
15.91
11.70
7
23.33
12.17
16
13.79
11.83
15 < Ne≤ 20
1
2.38
16.20
5
11.36
16.84
4
13.33
17.50
10
8.62
17.04
Ne> 20
0
—
—
4
9.09
21.00
2
6.67
21.00
6
5.17
21.00
总计Total
42
100.00
6.61
44
100.00
9.17
30
100.00
9.98
116
100.00
8.45
Range of Ne means the test location in one set of regional trial needed to achieve a heritability of 0.75 in half-open intervals, Ne stands for the test locations number needed to achieve a H = 0.75. 试点数范围是指在H=0.75时所需要的试验点数量的分布范围, 用Ne表示。
表3 2000-2014年全国棉花单年多点区试在遗传力为0.75时所需要试点数量的次数分布 Table 3 Frequency distribution of test location number needed at 0.75 of cross-trial heritability in the cotton regional trials from 2000 to 2014
表4 Table 4 表4(Table 4)
表4 全国棉花单年多点区试在不同遗传力水平下需要的试点数量定量分析 Table 4 Test location number needed to achieve different cross-location heritability levels for the regional cotton trials
棉区 Cotton region
噪信比 Qe
当前 试点数 N
不同遗传力(H)水平下需要的试点数 Test locations needed at different heritability (H) levels
0.55
0.65
0.75
0.85
0.90
0.91
0.92
0.93
0.94
0.95
长江流域 Yangtze River Valley
2.20
19
3
4
7
12
19
21
24
28
32
38
黄河流域 Yellow River Valley
3.06
20
4
6
9
17
27
30
33
38
45
53
西北内陆 Northwest Inland
3.33
9
4
7
10
19
29
32
36
42
48
58
全国 Nationwide
2.82
17
4
6
9
16
25
27
31
35
41
49
Qe is the noise-signal quotient at multi-location level regional trials. N is the current number of test locations. 噪信比(Qe)为单年多点区试的噪音与信号的比率; N为当前试点数为目前实际设置的试点数量。
表4 全国棉花单年多点区试在不同遗传力水平下需要的试点数量定量分析 Table 4 Test location number needed to achieve different cross-location heritability levels for the regional cotton trials
图1 全国棉花区域试验单年多点试验的遗传力(H)及其所需试点数量(Ne)的关系 YaRV、YeRV和NoWI分别代表长江流域、黄河流域和西北内陆棉区, Qe为各棉区区域试验的噪信比。Fig. 1 Relationship between heritability (H) and test location number (Ne) in one-year-multiple-sites cotton regional trials YaRV, YeRV, and NoWI stand for the Yangtze River Valley, the Yellow River Valley and the Northwest Inland cotton planting regions respectively. Qe is the noise-signal quotient.
4 结论我国棉花品种区域试验采用3次重复足以保证试验质量。长江流域和黄河流域国家棉花区试目前的试点数量设置已经可以充分满足试验的遗传力要求, 西北内陆棉区的试点数也基本上符合试验的遗传力要求。我国棉花区试中试点数量设置的优化方案为, 长江流域棉区保持当前20个左右的试点数量, 遗传力即可达到0.90的水平; 黄河流域和西北内陆棉区在国家财力、物力和人力等条件许可的情况下, 分别将试点数量增加到27个和19个左右, 遗传力可达到0.90和0.85的水平。 The authors have declared that no competing interests exist.
孔繁玲, 张群远, 杨付新, 郭恒敏. 棉花品种区域试验的精确度探讨. 作物学报, 1998, 24: 601-607Kong FL, Zhang QY, Yang FX, Guo HM. Studies on the precision of regional cotton variety trial. Acta Agron Sin, 1998, 24: 601-607 (in Chinese with English abstract)[本文引用:3]
[2]
翟虎渠. 应用数量遗传. 北京: 中国农业出版社, 2001Zhai H Q. Applied Quantitative Genetics. Beijing: China Agricultural Press, 2001 (in Chinese)[本文引用:1]
[3]
Yan WK. Crop Variety Trials: Data Management and Analysis. Oxford: Wiley-Blackwell, , 2014[本文引用:6]
[4]
Yan WK, Fregeau-ReidJ, MartinR, PageauD, Mitchell-FetchJ. How many test locations and replications are needed in crop variety trials for a target region?Euphytica, 2015, 202: 361-372[本文引用:5]
Yan WK. GGEbiplot—a windows application for graphical analysis of multienvironment trial data and other types of two-way data. Agron J, 2001, 93: 1111-1118[本文引用:1]
[7]
严威凯. 双标图分析在农作物品种多点试验中的应用. 作物学报, 2010, 36: 1805-1819Yan WK. Optimal use of biplots in analysis of multi-location variety test data. Acta Agron Sin, 2010, 36: 1805-1819 (in Chinese with English abstract)[本文引用:1]
[8]
张群远, 孔繁玲, 杨付新. 我国作物品种区域试验的精确研究. 中国农业大学学报, 2001, 6(1): 43-50Zhang QY, Kong FL, Yang FX. Evaluation of the precision of regional crop trials in China. J China Agric Univ, 2001, 6(1): 43-50 (in Chinese with English abstract)[本文引用:2]
[9]
金石桥, 许乃银. 我国棉花品种区域试验面临的挑战与对策. 中国棉花, 2012, 39(1): 12-14Jin SQ, Xu NY. The challenges and countermeasures in national-level cotton regional trials in China. China Cotton, 2012, 39(1): 12-14 (in Chinese )[本文引用:1]
[10]
许乃银, 金石桥. 棉花品种区域试验适宜试验点数量的抽样估计. 棉花学报, 2013, 25: 57-62Xu NY, Jin SQ. Sampling estimation of suitable quantity of test sites in cotton variety regional trials. Cotton Sci, 2013, 25: 57-62 (in Chinese with English abstract)[本文引用:1]
[11]
GouyM, RousselleY, BastianelliD, LecomteP, BonnalL, RoquesD, Efile JC, RocherS, DaugroisJ, ToubiL, NabenezaS, HervouetC, TelismartH, DenisM, Thong-ChaneA, Glaszmann JC, Hoarau JY, NiboucheS, CostetL. Experimental assessment of the accuracy of genomic selection in sugarcane. Theor Appl Genet, 2013, 126: 2575-2586[本文引用:1]
[12]
IwataH, JanninkJ. Accuracy of genomic selection prediction in barley breeding programs: a simulation study based on the real single nucleotide polymorphism data of barley breeding lines. Crop Sci, 2011, 51: 1887-1902[本文引用:1]
[13]
Kelly AM, Smith AB, Eccleston JA, Cullis, BR. The accuracy of varietal selection using factor analytic models for multi- environment plant breeding trials. Crop Sci, 2007, 47: 1063-1070[本文引用:1]
[14]
Gauch HG, Zobel RW. Accuracy and selection success in yield trial analyses. , 1989, 77: 473-481[本文引用:1]
[15]
王洁, 廖琴, 胡小军, 万建民. 北方稻区国家水稻品种区域试验精确度分析. 作物学报, 2010, 36: 1870-1876WangJ, LiaoQ, Hu XJ, Wan JM. Precision evaluation of rice variety regional trials in northern China. Acta Agron Sin, 2010, 36: 1870-1876 (in Chinese with English abstract)[本文引用:1]
[16]
Yan WK, Holland JB. A heritability-adjusted GGE biplot for test environment evaluation. Euphytica, 2010, 171: 355-369[本文引用:1]
[17]
许乃银, 李健. 棉花区试中品种多性状选择的理想试验环境鉴别. 作物学报, 2014, 40: 1936-1945Xu NY, LiJ. Identification of ideal test environments for multiple traits selection in cotton regional trials. Acta Agron Sin, 2014, 40: 1936-1945 (in Chinese with English abstract)[本文引用:1]
[18]
许乃银, 张国伟, 李健, 周治国. 基于HA-GGE双标图的长江流域棉花区域试验环境评价. 作物学报, 2012, 38: 2229-2236Xu NY, Zhang GW, LiJ, Zhou ZG. Evaluation of cotton regional trial environments based on HA-GGE biplot in the Yangtze River Valley. Acta Agron Sin, 2012, 38: 2229-2236 (in Chinese with English abstract)[本文引用:1]
[19]
Yan WK, Hunt LA, Sheng QL, SzlavnicsZ. Cultivar evaluation and mega-environment investigation based on the GGE biplot. Crop Sci, 2000, 40: 597-605[本文引用:1]
[20]
Yan WK. Mega-environment analysis and test-loction evaluation based on unbalanced multiyear data. Crop Sci, 2015, 55: 113-122[本文引用:1]
[21]
张志芬, 付晓峰, 刘俊青, 杨海顺. 用GGE双标图分析燕麦区域试验品系产量稳定性及试点代表性. 作物学报, 2010, 36: 1377-1385Zhang ZF, Fu XF, Liu JQ, Yang HS. Yield stability and testing-site representativeness in national regional trials for oat lines based on GGE-biplot analysis. Acta Agron Sin, 2010, 36: 1377-1385 (in Chinese with English abstract)[本文引用:1]
[22]
罗俊, 张华, 邓祖湖, 许莉萍, 徐良年, 袁照年, 阙友雄. 应用GGE双标图分析甘蔗品种(系)的产量和品质性状. 作物学报, 2013, 39: 142-152LuoJ, ZhangH, Deng ZH, Xu LP, Xu LN, Yuan ZN, Que YX. Analysis of yield and quality traits in sugarcane varieties (lines) with GGE-biplot. Acta Agron Sin, 2013, 39: 142-152 (in Chinese with English abstract)[本文引用:1]
[23]
罗俊, 许莉萍, 邱军, 张华, 袁照年, 邓祖湖, 陈如凯, 阙友雄. 基于HA-GGE双标图的甘蔗试验环境评价及品种生态区划分. 作物学报, 2015, 41: 214-227LuoJ, Xu LP, QiuJ, ZhangH, Yuan ZN, Deng ZH, Chen RK, Que YX. Evaluation of sugarcane test environments and ecological zone division in China based on HA-GGE biplot. Acta Agron Sin, 2015, 41: 214-227 (in Chinese with English abstract)[本文引用:1]
[24]
LaffontJ, HanafiM, WrightK. Numerical and graphical measures to facilitate the interpretation of GGE biplots. Crop Sci, 2007, 47: 990-996[本文引用:1]
[25]
ImtiazM, Malhotra RS, SinghM, ArslanS. Identifying high yielding, stable chickpea genotypes for spring sowing: specific adaptation to locations and sowing seasons in the Mediterranean region. Crop Sci, 2013, 53: 1472-1480[本文引用:1]
[26]
MohammadiR, AmriA. Genotype × environment interaction and genetic improvement for yield and yield stability of rainfed durum wheat in Iran. Euphytica, 2013, 192: 227-249[本文引用:1]
[27]
BaxevanosD, GoulasC, RossiJ, BraojosE. Separation of cotton cultivar testing sites based on representativeness and discriminating ability using GGE biplots. Agron J, 2008, 100: 1230-1236[本文引用:1]