删除或更新信息,请邮件至freekaoyan#163.com(#换成@)

基于主成分和SOM聚类分析的高粱品种萌发期抗旱性鉴定与分类

本站小编 Free考研考试/2021-12-26

王艺陶, 周宇飞, 李丰先, 依兵, 白薇, 闫彤, 许文娟, 高明超, 黄瑞冬*
沈阳农业大学农学院, 辽宁沈阳 110866
* 通讯作者(Corresponding author): 黄瑞冬, E-mail:r_huang@126.com
第一作者联系方式: E-mail:wangyitao-happy@163.com
收稿日期:2013-08-16 基金:本研究由国家现代农业产业技术体系建设专项(CARS-06-02-02)资助。

摘要采用人工气候箱内培养皿培养, PEG-6000溶液模拟干旱胁迫环境, 在萌发期以80、120、150和175 g L-1 PEG-6000水溶液处理31个高粱品种, 旨在根据高粱品种萌发期对不同干旱胁迫程度的响应, 筛选出具有抗旱能力的高粱品种并探讨高粱萌发期抗旱性鉴定的方法。通过主成分分析法(PCA)和神经网络自组织映射(SOM)聚类分析法对各高粱品种进行抗旱性综合分析与评定。PCA结果表明, 相对芽长、相对根长和相对萌发抗旱指数载荷量最大, 将其作为萌发期高粱抗旱性筛选的主要评价指标, 并对31个高粱品种抗旱性排序。通过SOM聚类分析将31个高粱品种按抗旱性强弱分为5类, 吉杂305等4个品种为高度抗旱品种, HL5等4个品种为抗旱品种, 辽杂10号等8个品种为中等抗旱品种, 锦杂103等7个品种为干旱敏感品种, 锦杂93等8个品种为高度干旱敏感品种。研究认为, 相对芽长、相对根长和相对萌发抗旱指数等可以作为高粱品种抗旱性鉴定的重要指标; SOM聚类分析可作为品种抗旱性分类的重要方法。

关键词:高粱; 抗旱性; 主成分分析; SOM聚类分析
Identification and Classification of Sorghum Cultivars for Drought Resistance during Germination Stage Based on Principal Components Analysis and Self Organizing Map Cluster Analysis
WANG Yi-Tao, ZHOU Yu-Fei, LI Feng-Xian, YI Bing, BAI Wei, YAN Tong, XU Wen-Juan, GAO Ming-Chao, HUANG Rui-Dong*
College of Agronomy, Shenyang Agricultural University, Shenyang 110866, China

AbstractPEG-6000 was used to simulate drought stress environment in artificial climate chamber for studying the effects of different concentrations of PEG-6000 (80, 120, 150, and 175 g L-1) on the response of 31 sorghum cultivars at germination stage, with the objectives of screening drought-resistant cultivars and exploring methods suitable for assessing drought-resistance of sorghum cultivars. Through principal component analysis (PCA) and Self-Organizing-Map (SOM) cluster analysis, the drought-resistance of 31 sorghum cultivars was evaluated. PCA showed that relative bud length, relative root length and drought-resistance index during germination had the maximum weight in values, and were used as the main indices for evaluating and ranking the drought-resistant capabilities of 31 sorghum cultivars. Five groups of drought-resistance capabilities of 31 sorghum cultivars were detected based on SOM cluster analysis: four cultivars including Jiza 305 were highly drought-resistant, four cultivars including HL8 were drought-resistant, eight cultivars including Liaoza 10 were medium drought-resistant, seven cultivars including Jinza 103 were drought-sensitive and eight cultivars including Jinza 93 were highly drought-sensitive. In conclusion, relative bud length, relative root length and germination drought-resistance index can be used as indices for identifying drought-resistance in sorghum, and SOM cluster analysis is an important method for classifying drought-resistance.

Keyword:Sorghum; Drought resistance; Principal component analysis (PCA); Self-Organizing-Map (SOM) cluster analysis
Show Figures
Show Figures




随着全球气候的变化, 干旱成为农业生产中经常遇到的问题, 是影响农业可持续发展的重要因素[ 1, 2]。充分利用抗旱性强的作物和品种以减少干旱所带来的不利影响, 对促进我国干旱、半干旱地区农业生产具有重要意义[ 3, 4, 5]。高粱是我国北方干旱、半干旱地区的重要作物之一。尽管其综合抗旱性较强, 但不同品种间抗旱性存在较大差异[ 6]。种子萌发阶段是作物能否在干旱条件下完成生育周期的关键时期之一, 萌发期抗旱性的强弱影响出苗好坏, 进而影响作物的生长发育。近年来, 许多****运用萌发胁迫指数法、灰色关联度法、隶属函数法和主成分分析法(PCA)等[ 7, 8, 9]对抗旱性进行了大量的研究, 但多数只侧重单一干旱处理, 具有一定的局限性, 不能反映出多水平干旱条件下作物的综合萌发状况。而近年来兴起的自组织映射(SOM)聚类分析可以较好
地从多水平、多因素着手, 建立多水平多观察值之间复杂的映射关系, 取得很好的分析效果, 已被广泛应用于各种聚类分析领域[ 10, 11]。本试验以4个不同PEG-6000溶液浓度模拟干旱胁迫, 运用PCA分析和SOM聚类分析法对31个高粱品种的萌发性状综合分析及评定, 为高粱抗旱的进一步研究和生产应用提供科学依据。
1 材料与方法1.1 供试材料31个高粱杂交品种分别来自辽宁省农业科学院、吉林省农业科学院、黑龙江省农业科学院、沈阳市农业科学院和锦州市农业科学院(表1)。
表1
Table 1
表1(Table 1)
表1 供试高粱品种及来源 Table 1 Sorghum cultivars tested and the origins
编号 Code品种 Cultivar来源 Origin编号 Code品种 Cultivar来源 Origin
S1辽杂10号 Liaoza 10辽宁省农业科学院 LnAASS17吉杂121 Jiza 121吉林省农业科学院 JlAAS
S2辽杂12 Liaoza 12辽宁省农业科学院 LnAASS18吉杂319 Jiza 319吉林省农业科学院 JlAAS
S3辽杂21 Liaoza 21辽宁省农业科学院 LnAASS19HL5黑龙江省农业科学院 HljAAS
S4锦杂93 Jinza 93锦州市农业科学院 JzAASS20HL8黑龙江省农业科学院 HljAAS
S5锦杂103 Jinza 103锦州市农业科学院 JzAASS21龙杂9号 Longza 9黑龙江省农业科学院 HljAAS
S6锦杂105 Jinza 105锦州市农业科学院 JzAASS22龙杂10号 Longza 10黑龙江省农业科学院 HljAAS
S7四杂25 Siza 25吉林省农业科学院 JlAASS23龙杂12 Longza 12黑龙江省农业科学院 HljAAS
S8四杂29 Siza 29吉林省农业科学院 JlAASS24辽杂11 Liaoza 11辽宁省农业科学院 LnAAS
S9吉杂90 Jiza 90吉林省农业科学院 JlAASS25沈杂5号 Shenza 5沈阳市农业科学院 SyAAS
S10吉杂97 Jiza 97吉林省农业科学院 JlAASS26辽杂19 Liaoza 19辽宁省农业科学院 LnAAS
S11吉杂123 Jiza 123吉林省农业科学院 JlAASS27辽杂35 Liaoza 35辽宁省农业科学院 LnAAS
S12吉杂124 Jiza 124吉林省农业科学院 JlAASS28龙杂11 Longza 11黑龙江省农业科学院 HljAAS
S13吉杂127 Jiza 127吉林省农业科学院 JlAASS29龙杂13 Longza 13黑龙江省农业科学院 HljAAS
S14吉杂305 Jiza 305吉林省农业科学院 JlAASS30锦杂106 Jinza 106锦州市农业科学院 JzAAS
S15辽杂15 Liaoza 15辽宁省农业科学院 LnAASS31葫梁1号 Huliang 1锦州市农业科学院 JzAAS
S16吉杂99 Jiza 99吉林省农业科学院 JlAAS
LnAAS: Liaoning Academy of Agricultural Sciences; JzAAS: Jinzhou Academy of Agricultural Sciences; JlAAS: Jilin Academy of Agricultural Sciences; HljAAS: Heilongjiang Academy of Agricultural Sciences; SyAAS: Shenyang Academy of Agricultural Sciences.

表1 供试高粱品种及来源 Table 1 Sorghum cultivars tested and the origins

1.2 试验设计从每品种选取25粒大小均匀一致、饱满的种子, 用7%的次氯酸钠溶液消毒10 min, 再用蒸馏水冲洗5次, 用滤纸吸干后摆放在直径9 cm并铺有2层滤纸的培养皿中, 种子间距离为粒长1~2倍。干旱胁迫液质量浓度分别为80、120、150和175 g L-1的PEG-6000水溶液, 与之对应的溶液水势分别为-0.1、-0.2、-0.3和-0.4 MPa[ 12], 对照以蒸馏水替代PEG-6000溶液, 每个培养皿中加入胁迫液8 mL, 每处理3次重复。将培养皿置RXZ-1000B型人工气候箱, 其昼/夜温度为28℃/25℃, 湿度为60%, 光照/黑暗为12 h/12 h, 光照强度为134 μmol m-2 s-1, 连续培养10 d。每2 d补充适量蒸馏水, 以保持胁迫液的水势不变。在第2、第4、第6、第8和第10天调查发芽数, 并于第10天测定芽长、根长、芽干重、根干重和种子剩余干重。
发芽势=第4天发芽种子粒数/供试种子粒数×100; 发芽率=第10天发芽种子粒数/供试种子粒数×100; 萌发抗旱指数(DGRI)=干旱胁迫下种子萌发指数(PIS)/对照种子萌发指数(PIC), 式中PIS/PIC = (1.00) nd2+(0.75) nd4+(0.50) nd6+(0.25) nd8, 其中, nd2 nd4 nd6 nd8分别为第2、第4、第6和第8天的萌发率[ 13]; 第10天, 从各处理随机取出高粱幼苗5株, 测定芽长和根长, 于85℃烘干后称量芽干重、根干重和籽粒剩余干重, 计算干物质转运率[ 14]
1.3 数据处理与分析用Microsoft Excel 2007整理数据并计算各处理性状的平均值和相对值。用SPSS 20.0进行相关性分析和PCA分析, 并对高粱品种的抗旱性进行综合评价。利用MATLAB 2012a对31个品种进行SOM分析[ 10]。为了减少各高粱品种间固有的差异, 对各测定指标均采用干旱胁迫处理和对照测定的相对值, 相对值比绝对值能更好地反映不同高粱品种的抗旱性[ 15, 16]

2 结果与分析2.1 不同高粱品种对萌发期干旱胁迫的响应由表2可知, 相对芽长、相对根长和相对萌发抗旱指数等8个测定指标在干旱胁迫下总体呈现不同程度的下降, 随着胁迫程度的加大, 测定指标较对照降幅增大。吉杂127 (S13)的芽长在-0.4 MPa水势下降幅最大, 相对芽长仅为0.33, 而处于同水势的HL5 (S19)为0.76, 降幅最小。龙杂9号(S21)在-0.1 MPa水势下具有较高的根长, 相对根长为1.21, 辽杂35 (S27)在-0.4 MPa水势下根长降幅最小, 相对根长为0.94。在-0.4 MPa水势下, 除吉杂127和龙杂11 (S28)相对发芽势分别为0.20和0.33外, 大多数品种相对发芽势在0.65~0.85之间, 平均值为0.74。4个水势下, 多数品种的相对发芽率在0.80~1.00之间, 龙杂9号在-0.1 MPa水势下相对发芽率最高, 为1.15, 吉杂127在-0.4 MPa水势下的相对发芽率最低, 为0.47。在-0.4 MPa水势下, 吉杂127相对芽干重和根干重均最低, 分别比对照下降65.0%和68.0%; 吉杂124 (S12)具有最高的相对芽干重, 龙杂10号(S22)具有最高的相对根干重, 二者分别仅比对照降低18.0%和8.0%。辽杂15 (S15)在-0.2 MPa水势下具有最高的相对干物质转运率。吉杂127在-0.4 MPa水势下的相对干物质转运率最低, 仅为0.26。HL8 (S20)在-0.1 MPa水势下的相对抗旱萌发指数最高, 为1.11, 而吉杂127最低, 为0.33, 其次为龙杂11 (S28), 其相对萌发抗旱指数比对照降低53.0%。上述结果表明, 低强度干旱胁迫(-0.1 MPa)对高粱种子萌发影响较小或略有提升活力作用; 高强度干旱胁迫(-0.34 ~ -0.40 MPa)足以使高粱不同抗旱性得到表现。
表2
Table 2
表2(Table 2)
表2 干旱胁迫条件下高粱各萌发性状的相对值 Table 2 Relative values of germination traits of sorghum cultivars under drought stress
品种 Cultivar水势 Water potential (MPa) X1 X2 X3 X4 X5 X6 X7 X8
S1-0.10.930.811.081.070.841.080.951.08
-0.20.890.740.980.990.830.990.790.97
-0.30.730.661.020.990.720.750.590.88
-0.40.700.530.940.890.600.580.550.75
S2-0.10.890.931.001.030.841.190.950.99
-0.20.720.741.050.970.730.830.680.90
-0.30.670.781.101.070.760.930.740.97
-0.40.520.490.710.850.540.660.480.57
S3-0.10.981.031.101.020.891.061.010.98
-0.20.860.700.930.910.840.860.870.89
-0.30.830.621.010.970.730.710.660.90
-0.40.690.500.870.790.680.770.590.71
S4-0.10.830.750.920.960.790.930.860.85
-0.20.780.710.850.900.770.860.760.74
-0.30.760.680.750.850.750.810.680.63
-0.40.580.530.710.810.620.740.600.56
S5-0.10.710.980.880.970.761.190.760.88
-0.20.680.740.890.960.690.990.690.83
-0.30.710.630.920.870.680.830.580.79
-0.40.590.560.760.840.590.650.500.63
S6-0.10.860.871.000.950.781.050.850.96
-0.20.800.751.000.950.780.910.800.95
-0.30.770.690.950.910.760.880.690.81
-0.40.680.590.890.900.650.750.580.69
S7-0.10.710.910.960.920.710.830.720.88
-0.20.570.680.920.940.550.690.540.88
-0.30.720.640.830.900.710.910.640.82
-0.40.450.470.730.890.450.550.410.75
S8-0.10.890.901.071.050.811.080.801.02
-0.20.470.700.910.870.520.850.540.84
-0.30.680.620.870.880.650.810.560.83
-0.40.580.580.760.780.580.750.480.72
S9-0.10.760.781.221.140.660.940.661.09
-0.20.470.850.970.980.521.080.580.93
-0.30.580.721.040.920.580.930.560.93
-0.40.390.610.700.820.440.860.450.71
S10-0.10.890.870.971.000.770.910.780.93
-0.20.550.721.010.960.570.740.580.93
-0.30.670.550.940.990.680.970.660.91
-0.40.600.470.750.890.550.800.510.78
S11-0.10.810.820.870.850.840.930.790.81
-0.20.500.640.950.960.640.770.610.91
-0.30.590.540.630.840.670.710.540.68
-0.40.600.430.640.770.640.620.520.68

表2 干旱胁迫条件下高粱各萌发性状的相对值 Table 2 Relative values of germination traits of sorghum cultivars under drought stress

表2-1
Table 2-1(Continued)
表2-1(Table 2-1(Continued))
表2-1 干旱胁迫条件下高粱各萌发性状的相对值 Table 2-1(Continued) Relative values of germination traits of sorghum cultivars under drought stress
品种 Cultivar水势 Water potential (MPa) X1 X2 X3 X4 X5 X6 X7 X8
S12-0.10.850.780.830.840.890.820.790.79
-0.20.710.670.930.930.790.600.610.90
-0.30.560.520.740.870.600.460.430.77
-0.40.600.540.640.840.820.560.520.68
S13-0.10.930.931.181.130.830.950.761.04
-0.20.670.730.770.760.630.760.580.70
-0.30.690.730.840.890.660.800.560.79
-0.40.330.320.200.470.350.320.260.33
S14-0.10.991.111.081.070.910.910.861.05
-0.20.981.001.021.000.860.880.871.04
-0.30.810.840.910.970.800.770.671.07
-0.40.440.350.650.910.450.650.360.97
S15-0.10.940.950.920.900.970.910.870.87
-0.20.850.800.900.900.910.951.010.85
-0.30.680.630.750.870.780.760.660.73
-0.40.700.580.800.850.770.730.740.75
S16-0.10.951.101.091.050.931.190.811.09
-0.20.940.911.060.940.891.080.840.91
-0.30.870.831.040.910.851.020.700.88
-0.40.720.801.000.950.740.880.620.89
S17-0.11.021.000.800.811.011.130.960.75
-0.20.880.930.870.850.831.080.780.79
-0.30.760.790.630.810.760.940.720.66
-0.40.670.690.590.740.620.790.500.61
S18-0.10.820.910.940.880.891.200.760.88
-0.20.660.610.760.830.680.900.540.76
-0.30.480.470.830.940.510.730.440.85
-0.40.380.300.640.660.350.530.550.62
S19-0.11.021.081.051.031.021.131.010.95
-0.21.021.020.910.870.991.090.870.78
-0.30.860.830.970.920.881.010.760.78
-0.40.760.710.910.910.690.820.590.73
S20-0.10.910.921.000.910.860.940.760.79
-0.20.870.821.060.980.780.930.770.94
-0.30.820.870.830.870.850.990.701.06
-0.40.690.630.830.800.610.770.521.11
S21-0.10.921.211.091.150.941.010.911.09
-0.20.900.941.211.130.840.900.801.08
-0.30.700.810.990.960.750.700.620.79
-0.40.490.631.021.090.520.490.440.95
S22-0.10.891.061.051.080.971.120.910.97
-0.20.670.731.000.930.830.940.690.83
-0.30.690.960.880.850.871.030.750.68
-0.40.630.860.830.830.770.880.640.71
S23-0.10.951.040.910.940.871.090.800.84
-0.20.811.000.940.930.791.100.740.84
-0.30.700.830.890.910.781.100.680.81
-0.40.610.700.700.870.630.920.500.67

表2-1 干旱胁迫条件下高粱各萌发性状的相对值 Table 2-1(Continued) Relative values of germination traits of sorghum cultivars under drought stress

表2-2
Table 2-2(Continued)
表2-2(Table 2-2(Continued))
表2-2 干旱胁迫条件下高粱各萌发性状的相对值 Table 2-2(Continued) Relative values of germination traits of sorghum cultivars under drought stress
品种 Cultivar水势 Water potential (MPa) X1 X2 X3 X4 X5 X6 X7 X8
S24-0.10.901.101.041.010.921.110.890.95
-0.20.761.001.080.950.721.050.610.94
-0.30.781.020.930.930.751.050.610.83
-0.40.620.610.900.930.550.700.400.84
S25-0.10.880.931.041.010.841.070.890.98
-0.20.771.021.020.990.851.030.850.96
-0.30.660.751.000.970.791.030.760.94
-0.40.600.680.930.930.700.880.620.89
S26-0.10.930.881.000.970.840.850.830.96
-0.20.820.820.790.940.870.930.770.85
-0.30.720.670.880.950.800.760.460.90
-0.40.650.630.650.910.750.750.590.81
S27-0.10.740.940.800.840.771.120.910.73
-0.20.770.850.830.810.830.750.720.73
-0.30.640.870.710.840.740.610.570.69
-0.40.660.850.740.790.730.600.580.70
S28-0.10.860.880.900.880.921.000.950.77
-0.20.720.870.850.890.880.990.920.76
-0.30.620.770.400.700.710.970.840.51
-0.40.640.770.330.690.730.850.720.47
S29-0.10.710.760.930.940.720.750.670.89
-0.20.650.870.880.880.711.110.630.77
-0.30.570.720.900.910.670.820.570.75
-0.40.620.800.800.850.690.800.590.70
S30-0.10.870.880.920.980.860.900.840.96
-0.20.830.920.851.010.810.980.820.94
-0.30.680.780.880.990.700.940.730.95
-0.40.630.670.710.890.620.730.650.79
S31-0.10.840.931.021.090.830.850.740.93
-0.20.771.010.860.950.770.970.690.74
-0.30.670.930.950.950.770.870.600.80
-0.40.600.800.620.940.580.630.480.70
X1: 相对芽长; X2: 相对根长; X3: 相对发芽势; X4: 相对发芽率; X5: 相对芽干重; X6: 相对根干重; X7: 相对干物质转运率; X8: 相对萌发抗旱指数。
X1:relative bud length; X2:relative root length; X3:relative germinating energy; X4: relative germination rate; X5: relative bud dry weight; X6: relative root dry weight; X7: relative dry matter transport rate; X8: relative germination drought-resistance index.

表2-2 干旱胁迫条件下高粱各萌发性状的相对值 Table 2-2(Continued) Relative values of germination traits of sorghum cultivars under drought stress

表3可知, 萌发期高粱品种间、同品种不同干旱胁迫条件下测定值指标间差异显著, 并且由此引起的互作差异显著, 说明干旱胁迫对高粱萌发性状影响明显, 不同基因型间存在差异。
表3
Table 3
表3(Table 3)
表3 高粱对干旱胁迫响应的方差分析 Table 3 ANOVA for response of sorghum to drought stress
性状 Trait方差结果 ANOVA results品种 Cultivar水势 Water potential品种×胁迫处理 Cultivar×water potential
X1 F29.65443.026.04
P<0.0001<0.0001<0.0001
X2 F132.241644.5823.33
P<0.0001<0.0001<0.0001
X3 F109.19690.9529.73
P<0.0001<0.0001<0.0001
X4 F40.2998.299.19
P<0.0001<0.0001<0.0001
X5 F668.612359.63478.62
P<0.0001<0.0001<0.0001
X6 F179.751682.17126.69
P<0.0001<0.0001<0.0001
X7 F119.712342.5924.25
P<0.0001<0.0001<0.0001
X8 F73.47399.1013.76
P<0.0001<0.0001<0.0001
X1-8 stand for germination traits.

表3 高粱对干旱胁迫响应的方差分析 Table 3 ANOVA for response of sorghum to drought stress

对31个高粱品种在4个胁迫浓度下的测定值均赋相同权重, 结果如
表4所示, 各萌发性状测定指标在干旱胁迫条件下几乎都小于对照, 不同品种降幅不同。相对芽长变化的范围为0.55~0.91, 相对根长的变化范围为0.58~0.99, 相对发芽势的变化范围为0.62~1.08, 相对发芽率的变化范围为0.83~1.12, 相对芽干重的变化范围为0.55~0.90, 相对根干重的变化范围为0.61~1.08, 相对干物质转运率的变化范围为0.51~0.86, 相对萌发抗旱指数的变化范围为0.63~1.03。吉杂305 (S14)和HL5 (S19)的芽长、根长、发芽势、发芽率、芽干重、根干重、干物质转运率和萌发抗旱指数相对值较大, 表现出较强的抗旱能力。吉杂123 (S11)和吉杂124 (S12)各项指标的相对值比较小, 对干旱胁迫比较敏感, 抗旱能力相对较差。干物质转运率可以反映出各品种对种子中贮藏物质的利用速率与效率以及体内能量供应水平[ 17]。龙杂11 (S28)的相对干物质转运率最高, 为0.86, 对干旱胁迫不敏感, 吉杂319 (S18)的相对干物质转运率最低, 为0.51, 对干旱胁迫较敏感。在干旱胁迫下, 萌发抗旱指数可避免因品种本身所具有的差异对试验结果的影响, 较客观地鉴定和评价品种对干旱胁迫的各种反应[ 18]。吉杂305 (S14)相对萌发抗旱指数最高, 为1.03, 而龙杂11 (S28)的相对萌发抗旱指数最低, 为0.63。
表4
Table 4
表4(Table 4)
表4 干旱胁迫条件下高粱各萌发性状权重后的相对值 Table 4 Relative values after weight of germination traits of sorghum cultivars under drought stress
品种 Cultivar X1 X2 X3 X4 X5 X6 X7 X8
S10.81 af0.68 ad1.01 ab1.02 ab0.75 ag0.85 ae0.72 ah0.92 ad
S20.76 ai0.74 ad0.96 ac1.06 ab0.72 bg0.90 ad0.71 ai0.86 bd
S30.84 ac0.71 ad0.98 ab0.96 ab0.79 ae0.85 ae0.78 ad0.87 bd
S40.74 bj0.67 bd0.81 ac0.96 ab0.73 bg0.84 ae0.72 ah0.69 cd
S50.67 ek0.73 ad0.87 ac0.98 ab0.68 dh0.92 ad0.63 dk0.78 bd
S60.78 ah0.73 ad0.96 ac0.98 ab0.74 ag0.90 ad0.73 ah0.85 bd
S70.62 ik0.67 bd0.86 ac0.97 ab0.60 gh0.75 ce0.58 hk0.84 bd
S80.68 dk0.70 ad0.90 ac0.91 ab0.64 eh0.87 ae0.59 fk0.85 bd
S90.55 k0.74 ad0.98 ab0.98 ab0.55 h0.95 ad0.56 ik0.91 ad
S100.68 dk0.65 cd0.92 ac0.95 ab0.65 eh0.86 ae0.63 dk0.89 ad
S110.63 hk0.61 d0.77 ac0.86 b0.70 cg0.76 be0.61 ek0.77 cd
S120.68 dk0.63 d0.78 ac0.87 b0.78 ae0.61 e0.59 gk0.78 bd
S130.66 fk0.67 cd0.74 ac0.83 b0.62 fh0.71 de0.54 jk0.71 bd
S140.81 af0.83 ad0.91 ab0.99 a0.75 af0.80 ae0.69 bj1.03 a
S150.79 ag0.74 ad0.84 ac0.95 ab0.86 ab0.84 ae0.82 ab0.80 bd
S160.87 ab0.91 ad1.05 ab1.01 ab0.85 ac1.04 ab0.74 ag0.94 ac
S170.83 ad0.85 ad0.72 bc0.83 b0.81 ad0.99 ad0.74 ag0.70 cd
S180.59 jk0.58 d0.79 ac0.88 ab0.61 gh0.83 ae0.51 k0.78 bd
S190.91 a0.91 ad0.96 ac1.00 ab0.90 a1.00 ac0.81 ac0.81 bd
S200.82 ae0.81 ad0.93 ac0.89 ab0.78 ae0.91 ad0.69 bj0.97 ac
S210.75 bi0.90 ad1.08 a1.12 ab0.76 af0.77 be0.69 bj1.01 ab
S220.72 bj0.99 ac0.94 ac0.97 ab0.85 ab1.08 ac0.75 af0.79 bd
S230.77 ai0.99 ac0.89 ac0.98 ab0.77 ae1.05 a0.68 bj0.81 bd
S240.77 ai0.93 ac0.99 ab1.01 ab0.74 bg0.98 ad0.63 dk0.89 ad
S250.73 bj0.85 ad1.00 ab0.98 ab0.79 ae1.00 ac0.78 ad0.94 ac
S260.78 ah0.75 ad0.82 ac0.98 ab0.82 ad0.82 ae0.66 ck0.87 bd
S270.70 cj0.88 ad0.77 ac0.88 ab0.77 af0.77 ae0.70 bj0.71 cd
S280.71 cj0.82 ad0.62 c0.89 ab0.81 ad0.96 ad0.86 a0.63 d
S290.64 gk0.79 ad0.88 ac0.96 ab0.70 cg0.87 ae0.62 ek0.78 bd
S300.75 bi0.81 ad0.84 ac1.01 ab0.75 ag0.89 ae0.76 ae0.91 ad
S310.72 bj0.92 ad0.86 ac1.01 ab0.74 bg0.83 ae0.63 dk0.79 bd
同一列值后相同字母表示0.05水平差异不显著。
Values followed by the same letter within the same column are not significantly different at the 0.05 probability level.

表4 干旱胁迫条件下高粱各萌发性状权重后的相对值 Table 4 Relative values after weight of germination traits of sorghum cultivars under drought stress

2.2 不同高粱品种萌发期各性状的相关性分析利用双变量Pearson简单相关系数法对干旱胁迫下萌发期8个指标性状的相对值进行相关性分析(表5)[ 19], 各性状之间均呈现正相关, 多数性状间的相关性达到了显著或极显著水平。相对萌发抗旱指数与相对发芽势和相对发芽率相关系数最大, 分别达到0.844**和0.822**。相对芽长与相对根长、相对芽干重和相对干物质转运率极显著正相关, 相关系数分别为0.494**、0.810**和0.716**。相对根长与相对芽干重和相对根干重也呈极显著正相关, 相关系数分别达到0.577**和0.731**
表5
Table 5
表5(Table 5)
表5 干旱胁迫下高粱各萌发性状的相关系数 Table 5 Correlation coefficients of germination traits of sorghum cultivars under drought stress
性状 Trait X1 X2 X3 X4 X5 X6 X7 X8
X11.000
X20.494**1.000
X30.382*0.3211.000
X40.403*0.3190.772**1.000
X50.810**0.577**0.1110.1831.000
X60.356*0.731**0.1760.1650.401*1.000
X70.716**0.418*0.0980.1880.807**0.398*1.000
X80.2910.1810.844**0.822**0.0260.0850.0061.000
***分别表示在0.05 和0.01的显著水平。
* and** indicate significance at the 0.05 and 0.01 probability levels, respectively. X1-8 stand for germination traits.

表5 干旱胁迫下高粱各萌发性状的相关系数 Table 5 Correlation coefficients of germination traits of sorghum cultivars under drought stress

2.3 不同高粱品种萌发期抗旱性主成分分析主成分分析是利用降维的思想, 描述隐藏在一组被测变量中无法直接测量到的隐性变量, 是综合处理多因素问题的有效方法[ 20]。本试验中, 不同胁迫强度下测得数据经加权平均后进行主成分分析, 所得结果如表6所示, 第i主成分的贡献率为47.674%。从主成分表达式可以看出, 在公式(1)中第i主成分上相对芽长和相对芽干重都有较高的载荷, 分别为0.429和0.385, 说明第i主成分由变量相对芽长决定的, 可以初步认定第i主成分反映高粱在干旱胁迫下的“芽部”萌发情况, 因此, 可以把第i主成分称之为“芽部性状因子”; 第ii主成分的贡献率为27.625%, 从公式(2)中可以看出, 第ii主成分在相对发芽势、相对发芽率和相对萌发抗旱指数上有较大的载荷, 反映的是高粱种子的萌发状况, 因此, 可以把第ii主成分称之为“萌发因子”; 第iii主成分的贡献率为12.457%, 在相对根长和相对根干重上有较大的载荷, 主要反映的是根部的生长发育状况, 因此, 可以把第iii主成分称之为“根部性状因子”。
表6
Table 6
表6(Table 6)
表6 3个主成分特征值及累计贡献率 Table 6 Eigen values of three principal components and their contribution and cumulative contribution rates
主成分 Principal component特征值 Eigen value贡献率 Contribution rate (%)累计贡献率 Cumulative contribution rate (%)
i3.81447.67447.674
ii2.21027.62575.299
iii0.99712.45787.756

表6 3个主成分特征值及累计贡献率 Table 6 Eigen values of three principal components and their contribution and cumulative contribution rates

前3个主成分的累计贡献率达87.756%, 三者足以说明该数据的变化趋势, 完全符合主成分分析的要求, 故取前3个主成分可以作为数据分析的有效成分。
经过主成分分析, 分离出的相互独立的8个主成分累计贡献率已达到100%, X1~ X8即可完全代表变量的所有变异程度。前3个主成分的累计贡献率达到87.756%, 可代表变量的绝大多数信息, 为了方便运算及形象展示, 可用前3个主成分描述变异。根据各品种的前3个主成分的因子载荷量绘图, 得到图1。各项指标由于不同的主成分载荷值而分布在不同的空间位置, 其中相对发芽势、相对发芽率和相对萌发抗旱指数之间的欧式距离较近, 说明他们之间在不同浓度处理下具有相似的响应。同样, 相对根长和相对根干重在不同浓度处理下具有相似的响应, 相对芽长、相对芽干重和相对干物质转运率在不同浓度处理下具有相似的响应。并且8个测定指标均不落在原点和坐标轴上, 说明这8个指标与这3个坐标轴所对应的因子都有关系。
图1
Fig. 1
Figure OptionViewDownloadNew Window
图1 干旱胁迫对高粱抗旱性影响前3个主成分散点图Fig. 1 Scatter plots of the first three principal components of drought-resistance under drought stress

表7表8分别为各成分的因子载荷矩阵和成分向量矩阵, 据此可获得各因子得分公式。
表7
Table 7
表7(Table 7)
表7 各因子载荷矩阵 Table 7 Loading matrix of each component
主成分 Principal component X1 X2 X3 X4 X5 X6 X7 X8
i0.8380.7440.6380.6700.7520.6000.6920.548
ii-0.222-0.2350.6840.633-0.524-0.307-0.5050.783
iii-0.3440.5050.018-0.065-0.2680.661-0.329-0.041

表7 各因子载荷矩阵 Table 7 Loading matrix of each component

表8
Table 8
表8(Table 8)
表8 各成分向量矩阵 Table 8 Components of the vector matrix
主成分 Principal component X1 X2 X3 X4 X5 X6 X7 X8
i0.4290.3810.3270.3430.3850.3070.3540.281
ii-0.149-0.1580.4600.426-0.352-0.207-0.3400.527
iii-0.3450.5060.018-0.065-0.2680.662-0.329-0.041

表8 各成分向量矩阵 Table 8 Components of the vector matrix

Y1=0.429 X1+0.381 X2+0.327 X3+0.343 X4+0.385 X5+0.307 X6+0.354 X7+0.281 X8 (1)
Y2=-0.149 X1-0.158 X2+0.460 X3+0.426 X4-0.352 X5-0.207 X6-0.340 X7+0.527 X8(2)
Y3=-0.345 X1+0.506 X2+0.018 X3-0.065 X4-0.268 X5+0.662 X6-0.329 X7-0.041×X8(3)
根据各品种 Y1 Y2 Y3的值, 各品种的综合得分公式为: Y=0.47674 Y1+0.27625 Y2+0.12457 Y3(4)
计算后 Y值见表9, Y值越大, 抗旱性越强。因此, 高粱品种萌发期抗旱性为S14> S16> S20> S21> S24> S19> S25> S23> S22> S1> S2> S3> S30> S6> S9> S27> S31> S26> S10> S15> S5> S29> S8> S17> S7> S4> S28> S18> S11> S12> S13。
表9
Table 9
表9(Table 9)
表9 各高粱品种主成分 Y值及抗旱性排序 Table 9 Principal component Y-values and the ranks of drought-resistance of sorghum cultivars
编号 Code Y1 Y1-value Y2 Y2-value Y3 Y3-value Y Y-value抗旱性排序 Order
S10.8300.979-1.2830.50610
S20.5280.794-0.5040.40811
S31.072-0.173-1.5580.26912
S4-1.065-1.006-0.943-0.90326
S5-1.1970.2590.349-0.45621
S60.4330.183-0.7480.16414
S7-2.5091.3830.149-0.79524
S8-1.6530.8420.353-0.51123
S9-1.9782.4201.741-0.05816
S10-1.3751.202-0.096-0.33520
S11-2.860-0.273-0.499-1.50130
S12-2.441-0.298-1.402-1.42128
S13-3.633-0.247-0.007-1.80131
S143.2212.9580.2432.3831
S150.651-1.380-1.603-0.27018
S163.586-0.2810.8911.7432
S17-0.040-2.865-0.198-0.83525
S18-3.7300.7590.572-1.49729
S193.083-1.647-0.5440.9476
S202.3722.124-0.6721.6343
S211.8082.138-0.5241.3874
S221.597-1.2390.9020.5318
S231.249-1.0901.7350.5109
S241.5090.4221.7661.0565
S251.6360.0830.2070.8297
S260.096-0.232-0.959-0.13817
S270.108-2.3692.243-0.32419
S28-0.432-3.673-0.121-1.23627
S29-1.3290.2570.609-0.48722
S300.637-0.041-0.5210.22713
S31-0.1730.0130.423-0.02615

表9 各高粱品种主成分 Y值及抗旱性排序 Table 9 Principal component Y-values and the ranks of drought-resistance of sorghum cultivars

2.4 不同高粱品种萌发期抗旱性聚类分析将31个高粱品种在4个不同PEG-6000溶液浓度下测得的8个性状的相对值数据标准化后, 利用数据挖掘软件MATLAB进行自组织映射(SOM)聚类分析[ 21], 将标准化的数据作为网络的输入层, 进行100 000次的训练后稳定, 得到聚类分析结果如表10所示。
聚类分析结果表明, 31个高粱品种可以分为5种类型, 同时根据主成分分析结果判断, 从第I类到第V类基本遵循抗旱性由弱到强排列。第I类高粱品种抗旱性最弱, 属于高度干旱敏感品种, 含锦杂93 (S4)、四杂25 (S7)和吉杂123 (S11)等8个品种; 第II类高粱品种抗旱性较弱, 属于干旱敏感品种, 含锦杂103 (S5)、四杂29 (S8)和吉杂97 (S10)等7个品种; 第III类高粱品种抗旱性一般, 属于中等抗旱性品种, 含辽杂10号(S1)、辽杂12 (S2)和辽杂21(S3)等8个品种; 第IV类高粱品种抗旱性较强, 属于抗旱性品种, 含HL5(S19)、龙杂10号(S22)和辽杂11 (S24)等4个品种; 第V类高粱品种抗旱性最强, 属于高度抗旱性品种, 含吉杂305 (S14)、吉杂99 (S16)和HL8 (S20)等4个品种。
表10
Table 10
表10(Table 10)
表10 基于SOM神经网络的聚类分析结果 Table 10 Classification of drought-resistance of sorghum cultivars based on SOM neural network
抗旱类别 Type of drought-resistance品种编号 Cultivar code
I: 高度干旱敏感 Highly drought-sensitiveS4, S7, S11, S12, S13, S17, S18, S28
II: 干旱敏感 Drought-sensitiveS5, S8, S10, S15, S26, S27, S29
III: 中等抗旱 Medium drought-resistantS1, S2, S3, S6, S9, S23, S30, S31
IV: 抗旱 Drought-resistantS19, S22, S24, S25
V: 高度抗旱 Highly drought-resistantS14, S16, S20, S21

表10 基于SOM神经网络的聚类分析结果 Table 10 Classification of drought-resistance of sorghum cultivars based on SOM neural network


3 讨论3.1 高粱品种萌发期抗旱性的综合评价植物在长期进化中形成各种抗旱机能, 并表现出不同生育阶段的差异性。萌发期是植物可能遇到的首个干旱胁迫时期, 是实现苗全、苗齐、苗壮的关键时期。通过简便、快速而准确的萌发期抗旱性鉴定分析方法, 筛选出具有萌发期抗旱性的作物品种, 具有重要理论意义和实践价值。多种作物上的抗旱性研究都表明, 作物抗旱性受多种因素影响, 以单一指标评价高粱萌发期的抗旱能力具有一定的片面性, 而多指标评价又过于繁琐和低效, 通过其中的几个主要因素进行分析则可达到快速、准确的目的[ 22, 23]。因此, 本研究利用了主成分和SOM聚类分析的方法, 对31个高粱品种萌发期抗旱性进行了鉴定与分类, 并取得了比较符合实际的研究结果。主成分分析法中, 将4个胁迫浓度下测得的数据均设权重为1, 在计算中涉及到品种或测定指标的相关系数的情况下, 可能会忽略一些高粱品种在不同胁迫程度下的特征差异, 但从抗旱性的排序和分类结果上看, 与试验中的观察结果较为一致, 能够有效地对31个高粱品种的抗旱性进行评定。
本试验中, 干旱胁迫下, 大多数高粱种子芽长、根长、发芽势、发芽率、芽干重、根干重、干物质转运率和萌发抗旱指数都不同程度地下降, 说明干旱胁迫抑制了高粱萌发的生理过程, 但轻度胁迫条件下个别品种某些测定指标较对照有所升高, 可能是一定程度上干旱胁迫促进了种子的某些萌发进程, 这与前人研究的适当逆境胁迫可能会提高种子发芽率和活力具有相似之处[ 24, 25]。作物的抗旱性主要表现在对逆境的适应性和敏感性上, 强大的根系是其适应干旱胁迫的重要特征之一。本试验结果表明, 在萌发期干旱胁迫条件下, 根的生长状况与芽长显著正相关。通过主成分分析可知, 芽部性状因子对高粱萌发期抗旱性影响最大, 因此, 在田间鉴定不便于测定根系时, 也可主要通过对芽部性状因子的研究评价高粱的抗旱性。
3.2 自组织映射(SOM)聚类分析结果的判定自组织映射(SOM)神经网络是无导师自组织和自学习的网络, 具有自稳定性。SOM聚类过程无需外界定义出评价函数, 能够识别出向量空间中最有意义的特征, 将高维数据压缩到二维平面, 可以广泛应用于数据的聚类分析, 并且这一过程无需人为干涉赋予各因子权重值, 从而简化分类过程, 得到较稳定的分类结果。本试验中, SOM分类结果与主成分分析结果较为一致, 说明SOM应用于高粱抗旱性分类是可行的。SOM的缺点是压缩过程中难免有数据信息损失[ 26, 27, 28, 29]。与单一胁迫水平的聚类分析不同, 本研究中聚类分析全面考虑4个胁迫水平上31个高粱品种萌发特征的综合表现, 涉及到的数据量很大, 少量的信息损失不影响结果判定的准确性。实际应用中, 如何避免神经网络训练时出现“死神经元”造成数据损失, 以及如何在神经网络训练时确定较少的训练次数以节约时间并得出较为稳定的输出结果, 需要试验人员根据实际情况判别和遴选, 但是SOM神经网络聚类分析不失为种质资源抗旱性鉴定的一个较为简便、理想的方法。

4 结论通过主成分分析和SOM聚类分析, 将31个高粱品种按照抗旱性强弱分为5大类, 吉杂305和吉杂99等4个品种为高度抗旱品种, HL5和龙杂10号等4个品种为抗旱品种, 辽杂10号和辽杂12等8个品种为中等抗旱品种, 锦杂103和四杂29等7个品种为干旱敏感品种, 锦杂93和四杂25等8个品种为高度干旱敏感品种。相对芽长、相对根长和相对萌发抗旱指数可以作为高粱品种抗旱性鉴定的重要指标。
The authors have declared that no competing interests exist.
作者已声明无竞争性利益关系。

参考文献View Option
原文顺序
文献年度倒序
文中引用次数倒序
被引期刊影响因子

[1]Lobell D B, Field C B. Global scale climate-crop yield relationships and the impacts of recent warming. Environ Res Lett, 2007, 2: 1-7[本文引用:1][JCR: 3.631]
[2]Hlavinka P, Trnka M, Semeradova D, Dubrovsky M, Zalud Z, Mozny M. Effect of drought on yield variability of key crops in Czech Republic. Agric Yor Meteorol, 2009, 149: 431-442[本文引用:1]
[3]山仑, 陈培元. 旱地农业生理生态基础. 北京: 科学技术出版社, 1998. pp1-18
Shan L, Chen P Y. Dryland agriculture ecological and physiological basis. Beijing: Science Press, 1998. pp1-18(in Chinese)[本文引用:1]
[4]崔江慧, 李霄, 常金华. PEG模拟干旱胁迫对高粱幼苗生理特性的影响. 中国农学通报, 2011, 27(9): 160-165
Cui J H, Li X, Chang J H. Biochemical characteristics in sorghum seedlings under PEG simulated drought stress. Chin Agric Sci Bull, 2011, 27(9): 160-165 (in Chinese with English abstract)[本文引用:1]
[5]Li Y P, Ye W, Wang M, Yan X D. Climate change and drought: a risk assessment of crop-yield impacts. Clim Res, 2009, 39: 31-46[本文引用:1][JCR: 1.994]
[6]山仑, 徐炳成. 论高粱的抗旱性及在旱区农业中的地位. 中国农业科学, 2009, 42: 2342-2348
Shan L, Xu B C. Discussion on drought resistance of sorghum and its status in agriculture in arid and semiarid regions. Sci Agric Sin, 2009, 42: 2342-2348 (in Chinese with English abstract)
[本文引用:1][CJCR: 1.4522]
[7]张智猛, 万书波, 戴良香, 吴正峰, 陈静, 苗华荣. 花生萌芽期水分胁迫品种适应性及抗旱性评价. 干旱地区农业研究, 2009, 27(5): 173-182
Zhang Z M, Wan S B, Dai L X, Wu Z F, Chen J, Miao H R. A study on appraisal of suitability and drought-resistance of different peanut varieties at germination stage. Agric Res Arid Areas, 2009, 27(5): 173-182 (in Chinese with English abstract)[本文引用:1]
[8]孟庆立, 关周博, 冯佰利, 柴岩, 胡银岗. 谷子抗旱相关性状的主成分与模糊聚类分析. 中国农业科学, 2009, 42: 2667-2675
Meng Q L, Guan Z B, Feng B L, Chai Y, Hu Y G. Principal component analysis and fuzzy clustering on drought-tolerance related traits of foxtail millet (Setaria italica). Sci Agric Sin, 2009, 42: 2667-2675 (in Chinese with English abstract)
[本文引用:1][CJCR: 1.4522]
[9]李文娆, 张岁岐, 山仑. 水分胁迫下紫花苜蓿和高粱种子萌发特性及幼苗耐旱性. 生态学报, 2009, 29: 3066-3074
Li W R, Zhang S Q, Shan L. Seeds germination characteristics and drought-tolerance of alfalfa and sorghum seedling under water stress. Acta Ecol Sin, 2009, 29: 3066-3074 (in Chinese with English abstract)[本文引用:1]
[10]刘鑫, 迟道才, 吴萍. 基于MATLAB的SOM网络的干旱聚类分析. 沈阳农业大学学报, 2008, 39: 61-64
Liu X, Chi D C, Wu P. SOM network’s on clustering analysis of the drought based on Matlab. J Shenyang Agric Univ, 2008, 39: 61-64 (in Chinese with English abstract)[本文引用:2][CJCR: 0.5722]
[11]文勇立, 李学东, 王永, 马力, 章杰, 陈世忠, 杨克美. 种养结合模式下基于R平台的土壤养分SOM神经网络分析. 西南民族大学学报(自然科学版), 2011, 37: 393-400
Wen Y L, Li X D, Wang Y, Ma L, Zhang J, Chen S Z, Yang K M. Self-organizing mapping (SOM) neural network analysis of soil nutrient based on R after using the model of combination of planting and breeding. J Southwest Univ Natl (Nat Sci Edn), 2011, 37: 393-400 (in Chinese with English abstract)[本文引用:1]
[12]Michel B E, Kaufmann M R. The osmotic potential of polyethylene glycol 6000. Plant Physiol, 1973, 51: 914-916[本文引用:1][JCR: 6.535]
[13]付学琴, 贺浩华, 文飘, 罗向东, 谢建坤. 东乡野生稻回交重组系的抗旱性评价体系. 应用生态学报, 2012, 23: 1277-1285
Fu X Q, He H H, Wen P, Luo X D, Xie J K. Drought resistance evaluation system for backcross lines of Dongxiang common wild rice (Oryza rufipogon Griff. ). Chin J Appl Ecol, 2012, 23: 1277-1285 (in Chinese with English abstract)
[本文引用:1][CJCR: 1.7261]
[14]齐华, 许晶, 孟显华, 王晓波, 刘明, 管宏军. 水分胁迫下燕麦萌芽期抗旱指标的研究. 种子, 2009, 28(7): 7-10
Qi H, Xu J, Meng X H, Wang X B, Liu M, Guan H J. Studies on the drought resistance index of oat at germination stage under water stress. Seed, 2009, 28(7): 7-10 (in Chinese with English abstract)[本文引用:1][CJCR: 0.5051]
[15]孙璐, 周宇飞, 汪澈, 肖木辑, 陶冶, 许文娟, 黄瑞冬. 高粱品种萌发期耐盐性筛选与鉴定. 中国农业科学, 2012, 45: 1714-1722
Sun L, Zhou Y F, Wang C, Xiao M J, Tao Y, Xu W J, Huang R D. Screening and identification of sorghum cultivars for salinity to- lerance during germination. Sci Agric Sin, 2012, 45: 1714-1722 (in Chinese with English abstract)
[本文引用:1][CJCR: 1.4522]
[16]田伯红, 王素英, 李雅静, 王建广, 张立新, 梁凤芹, 翟玉柱, 刘金荣. 谷子地方品种发芽期和苗期对NaCl胁迫的反应和耐盐品种筛选. 作物学报, 2008, 34: 2218-2222
Tian B H, Wang S Y, Li J Y, Wang J G, Zhang L X, Liang F Q, Zhai Y Z, Liu J R. Response to sodium chloride stress at germination and seedling and identification of salinity tolerant genotypes in foxtail millet land races originated from China. Acta Agron Sin, 2008, 34: 2218-2222 (in Chinese with English abstract)
[本文引用:1][CJCR: 1.8267]
[17]安永平, 强爱玲, 张媛媛, 张文银, 曹桂兰, 韩龙植. 渗透胁迫下水稻种子萌发特性及抗旱性鉴定指标研究. 植物遗传资源学报, 2006, 7: 421-426
An Y P, Qiang A L, Zhang Y Y, Zhang W Y, Cao G L, Han L Z. Study on characteristics of germination and drought-resistance index by osmotic stress in rice. J Plant Genet Resour, 2006, 7: 421-426 (in Chinese with English abstract)[本文引用:1][CJCR: 1.1628]
[18]时立文. SPSS19. 0统计分析从入门到精通. 北京: 清华大学出版社, 2012. pp300-305
Shi L W. SPSS19. 0 statistical analysis savvy. Beijing: Tsinghua University Press, 2012. pp300-305(in Chinese)[本文引用:1]
[19]高福元, 赵成章, 卓玛兰草. 高寒山地甘肃臭草斑块特征与土壤水分的关系. 应用与环境生物学报, 2012, 18: 571-574
Gao F Y, Zhao C Z, Zhuo Malancao. Relationship between soil moisture and patch characteristics of Melica przewalskyii communities in a degraded alpine grassland in Gansu, China. Chin J Appl Environ Biol, 2012, 18: 571-574 (in Chinese with English abstract)[本文引用:1][CJCR: 0.8262]
[20]朱春燕, 黄丹枫, 蔡保松, 俞晨玺, 邓浩宣, 卓俊义. 甜瓜品种资源萌发期耐盐性及其指标评价. 上海交通大学学报(农业科学版), 2010, 28: 504-508
Zhu C Y, Huang D F, Cai B S, Yu C X, Deng H X, Zhuo J Y. Salt tolerance and indicators assessment of melon germplasms in germination stage. J Shanghai Jiaotong Univ (Agric Sci Edn), 2010, 28: 504-508 (in Chinese with English abstract)[本文引用:1]
[21]孙景宽, 张文辉, 张洁明, 刘宝玉, 刘新成. 种子萌发期4种植物对干旱胁迫的响应及其抗旱性评价研究. 西北植物学报, 2006, 26: 1811-1818
Sun J K, Zhang W H, Zhang J M, Liu B Y, Liu X C. Response to droughty stresses and drought-resistances evaluation of four species during seed germination. Acta Bot Boreali-Occident Sin, 2006, 26: 1811-1818 (in Chinese with English abstract)[本文引用:1][CJCR: 1.1098]
[22]马斌, 周志宇, 张丽莉, 高文星. 阿拉善荒漠区土壤因子与白沙蒿人工种群生长的主成分分析. 西北植物学报, 2007, 27: 859-863
Ma B, Zhou Z Y, Zhang L L, Gao W X. The principal component analysis of soil and population growth status of Artemisia sphaerocephala in arid region of Alex desert. Acta Bot Boreali-Occident Sin, 2007, 27: 859-863 (in Chinese with English abstract)[本文引用:1][CJCR: 1.1098]
[23]李翠, 梁燕, 张纪涛, 罗雷. 渗透胁迫对番茄种子萌发特性的影响. 干旱地区农业研究, 2011, 29: 173-179
Li C, Liang Y, Zhang J T, Luo L. Effects of osmotic stress on germination characteristics of tomato. Agric Res Arid Areas, 2011, 29: 173-179 (in Chinese with English abstract)[本文引用:1]
[24]罗音, 王玉军, 谢胜利, 赵新西, 杨兴洪, 王玮. 等渗水分与盐分胁迫对烟草种子萌发的影响及外源甜菜碱的保护作用. 作物学报, 2005, 31: 1029-1034
Luo Y, Wang Y J, Xie S L, Zhao X X, Yang X H, Wang W. Effects of isotonic water and salt stress on seed germination of tobacco and protective function of exogenous glycinebetaine. Acta Agron Sin, 2005, 31: 1029-1034 (in Chinese with English abstract)
[本文引用:1][CJCR: 1.8267]
[25]刘杰, 刘公社, 齐冬梅, 李芳芳, 汪恩华. 聚乙二醇处理对羊草种子萌发及活性氧代谢的影响. 草业学报, 2002, 11: 59-64
Liu J, Liu G S, Qi D M, Li F F, Wang E H. Effect of PEG on germination and active oxygen metabolism in wildrye (Leymus chinensis) seeds. Acta Pratac Sin, 2002, 11: 59-64 (in Chinese with English abstract)[本文引用:1]
[26]陈家伟, 陈清华, 尹春华, 方福康. SOM与中国不同地区人力资本构成分析. 北京师范大学学报(自然科学版), 2006, 42: 107-110
Chen J W, Chen Q H, Yin C H, Fang F K. SOM and Chinese human capital components analysis between areas. J Beijing Norm Univ (Nat Sci), 2006, 42: 107-110 (in Chinese with English abstract)[本文引用:1]
[27]吴春旭, 鲍满园, 苟清龙. 自组织映射聚类算法在电信客户细分中的应用. 计算机系统应用, 2010, 19(8): 168-172
Wu C X, Bao M Y, Gou Q L. Application of self organization map to classification of the telecommunication company. Comp Syst Appl, 2010, 19(8): 168-172 (in Chinese with English abstract)[本文引用:1]
[28]陈志强, 陈健飞. 基于SOM的漳浦样区土壤分类研究. 吉林师范大学学报(自然科学版), 2007, (2): 4-6
Chen Z Q, Chen J F. The soil classification study based on SOM in Zhangpu. Jilin Norm Univ J (Nat Sci Edn), 2007, (2): 4-6 (in Chinese with English abstract)[本文引用:1]
[29]Lu Y L, Hao Z F, Xie C X. Large-scale screening for maize drought resistance using multiple selection criteria evaluated under water-stressed and well-watered environments. Field Crops Res, 2011, 124: 37-45[本文引用:1][JCR: 2.474]
相关话题/农业 种子 指标 物质 作物