Developing herbicide resistant Sri-lankan rice (Oryza sativa L.) varieties: An application of Self Organizing Map

Accepted Manuscript Developing Herbicide Resistant Sri-lankan Rice (Oryza sativa L.) Varieties: an Application of SELF Organizing Map E.M.S.I. Ekanyaka, S.R. Weerakoon, S. Somaratne, O.V.D.S.J Weerasena PII: DOI: Reference:

S2214-3173(16)30078-6 http://dx.doi.org/10.1016/j.inpa.2017.02.002 INPA 75

To appear in:

Information Processing in Agriculture

Received Date: Accepted Date:

10 August 2016 13 February 2017

Please cite this article as: E.M.S.I. Ekanyaka, S.R. Weerakoon, S. Somaratne, O.V.D.S.J Weerasena, Developing Herbicide Resistant Sri-lankan Rice (Oryza sativa L.) Varieties: an Application of SELF Organizing Map, Information Processing in Agriculture (2017), doi: http://dx.doi.org/10.1016/j.inpa.2017.02.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

DEVELOPING HERBICIDE RESISTANT SRI-LANKAN RICE

2

(Oryza sativa L.) VARIETIES: AN APPLICATION OF SELF ORGANIZING MAP

3

E.M.S.I. Ekanyaka1, 2, S. R. Weerakoon1*, S. Somaratne1, O.V.D.S.J Weerasena2

4

1

5

2

Department of Botany, The Open University of Sri Lanka, P.O. Box 21, Nawala, Sri Lanka Institutes of Biochemistry, Molecular Biology and Biotechnology, University of Colombo

6 7

Corresponding author; E- Mail: [email protected]

8

TEL: +94 714933922

9

FAX: +94 112806577

10 11

ABSTRACT

12

Application of high concentrations of post-emergent broad spectrum systemic herbicide, glyphosate is

13

prevalently used to control rice weeds in Asian countries including Sri Lanka. Off target movements of

14

glyphosate adversely affect growth of the rice plant reducing the yield. Inducing herbicide resistance

15

(HR) in cultivated rice is a novel approach to enhance selectivity and crop safety. Studies on induced HR

16

in Sri Lankan rice varieties are limited and studies are required to include HR rice in a cropping program.

17

Ethyl Methyl Sulfonate (EMS), a chemical mutagen is used for functional mutations. The present study

18

is an attempt of raising HR rice lines through conventional breeding methods. A suitable glyphosate

19

concentration was determined by a preliminary study using five rice varieties (Bg300, Bg352, At362,

20

Bg379-2 and H4) and varying glyphosate concentrations (0.25g/l, 0.5g/l, 1 g/l, 2g/l and 3g/l). Resistance

21

percentage ≥ 50% was arbitrary considered as resistant varieties. Complete Randomized Design with

22

three replicates was used in the experiment. Agro-morphological characters were recorded for the

23

survived plants. Twelve varieties (Bg359, At362, Bw364, Ld365, Bg366, Bg369, Bg379-2, Bg403,

24

Bg454 “Pachcha perumal”, “Kalu heenati” and “Kurulu thuda”) showed natural resistance to glyphosate

25

and fourteen varieties have increased resistance after mutagenesis by EMS (S0 – First generation). HR

26

resistance percentage of S1 (Second generation) plants was similar to S0, plants indicating HR was

27

transferred to new generation. Conventional statistics was supplemented with Self-Organizing Maps

28

(SOM) to visualize variation of agro-morphological characters under treatment. The result proved that

29

EMS application is an effective method in breeding new rice germplasm with HR and SOM is an

30

important tool for visualizing the multi-dimensional dataset in lower dimensions.

31

Key words: Ethyl Methyl Sulfonate, glyphosate, Herbicide Resistance, Oryza sativa, Self-Organizing

32

Map 1

33 34

Author summary

35

E.M.S.I. Ekanayaka is a post graduate student working on Herbicide resistant rice in Sri Lanka at the

36

Institutes of Biochemistry, Molecular Biology and Biotechnology, University of Colombo, Sri Lanka.

37

Prof. S. R. Weerakoon is a Professor in Botany at the Department of Botany, The Open University of Sri

38

Lanka, specialized in Genetics, Plant Breeding and Biotechnology, currently involved in studies of

39

Herbicide resistant rice and weed management in Sri Lanka. Dr. S. Somaratne is a Senior Lecture in

40

Botany, Department of Botany, Open University of Sri Lanka currently involve in number of multi-

41

disciplinary researches in Sri Lanka. Further, he involved in statistical analysis of data in the present

42

paper. Dr. O.V.D. S. J. Weerasena is a Senior Lecture in Molecular Biology, Institutes of Biochemistry,

43

Molecular Biology and Biotechnology, University of Colombo, currently involved in molecular

44

characterization of plants and insects.

45 46

INTRODUCTION

47

Weeds cause serious yield reduction (up to 9.5%) in rice production worldwide [1]. Sri Lanka, a country

48

which produces sufficient amount of rice, encounter a problem of weeds as the major biotic stress in rice

49

production leading to a yield loss in the range of 30% to 40 % [2]. Therefore, weed management

50

techniques are necessary to obtain a good yield because weeds are so prevalent in almost all the areas of

51

rice cultivation. The common methods used to manage weeds include cultural, mechanical, biological, and

52

chemical means. An integrated approach using a variety of methods in combination provides most

53

successful weed management in rice fields [3]. Weed control using herbicides is the most popular method

54

among farmers and it allows economically viable weed control providing cost-effective method in the

55

production of agricultural crops [4].

56

Glyphosate is a non-selective, foliar, post-emergence herbicide used to control annual and perennial

57

weeds in pre-plant burn down applications and for weed control in glyphosate–resistant crops [5].

58

Glyphosate has been used as the most effective herbicide to control weeds in rice cultivation areas in the

59

world. Off target movements of glyphosate causes damages to the cultivated rice by reducing the yield

60

up to 80% [6; 7]. Thus, there is need in developing crop plants which are not affected by the broad

61

spectrum herbicides. The crop plants less affected with broad spectrum herbicides are generally referred

62

to as herbicide resistant plants. The term herbicide resistance is defined in various ways. The official 2

63

Weed science society of America (WSSA) definitions of “herbicide resistance” and “herbicide tolerance”

64

were adopted in the present research. “Herbicide resistance is the inherited ability of a plant to survive

65

and reproduce following exposure to a dose of herbicide normally lethal to the wild type. In a plant,

66

resistance may be naturally occurring or induced by such techniques as genetic engineering or selection

67

of variants produced by tissue culture or mutagenesis.”

68

“Herbicide tolerance is the inherent ability of a species to survive and reproduce after herbicide

69

treatment. This implies that there was no selection or genetic manipulation to make the plant tolerant; it is

70

naturally tolerant” [8].

71

Conventional plant breeding techniques enhance the development of herbicide resistant (HR) crops over

72

the last two decades. Ethyl Methyl Sulfonate (EMS) is the most commonly used chemical mutagen in

73

plants which would provide a sequence of change of functional mutation. N-(phosphonomethyl) glycine

74

commercially known as glyphosate targets the enzyme 5-pyruvylshikimate 3-phosphate synthase

75

(EPSPS) which involved in biosynthesis of aromatic amino acids [9]. Mutants of EPSPS are resulted in

76

glyphosate tolerant crops [10].

77

Inducing HR into rice is a new means to confer selectivity and enhance crop safety and production [11].

78

HR crops provide additional crop choice, enabling implementation of alternate weed management tactics

79

to target specific weeds while maintaining crop sequences. Rice is the staple food of more than half of the

80

world’s population as well as in Sri Lanka. Therefore, it is very important to carry out studies on HR

81

varieties because farmers are continuously applying a massive amount of herbicides on rice crops to

82

reduce the noxious effects of weed populations on the growth performance of rice. It is believed that

83

introduction of an HR crop in a cropping program along with a range of weed management tactics ensure

84

controlling of hard-to-control weeds [12]. Evaluation of glyphosate resistance among Sri Lankan Oryza

85

species is important to identify the genes conferring HR. Therefore, it is worthwhile to screen the Sri

86

Lankan rice (Oryza spp.) gene pool for HR and consider the possibility of incorporating it in rice-

87

breeding programs. Present study was conducted to evaluate variability of natural resistance among Sri

88

Lankan cultivated rice varieties (inbred and traditional) for glyphosate and attempts were made to

89

produce stable resistant lines by applying conventional mutation breeding methods instead of genetic

90

engineering methods.

While Herbicide tolerance is defined as

91 92 93

METHODS

94 3

95

Materials - Rice varieties

96

Bg94-1, Bg250, Bg300, Bg304, Bg305, Bg352, Bg357, Bg359, Bg360, At362, Bw364, Ld365, Bg366,

97

Bg369, Bg379-2, H4, Bg403, Bg450, Bg454 inbred-developed (cultivated) rice varieties and Five

98

traditionally cultivated varieties (“Kalu heenati”, “Kurulu thuda”, “Suwadal”, “Rathhal”, “Madel”, and

99

“Pachchaperumal”) were collected from Rice Research Development Institute at Batalagoda,

100

Ambalanthota, Bombuwela and Labuduwa, Sri Lanka for the study.

101 102 103

Method 1

104

A study was carried out in a plant house at The Open University, Nawala in the Low-Country Wet Zone,

105

in the Western Province of Sri Lanka from September 2013 to March 2015. There was an average

106

ambient temperature of 28 -32°C and average relative humidity of 80-85% in the experiment site during

107

the experiment period. The puddle soil was collected from paddy field and filled in to pots (5 Kg/pot).

108

The selected seeds were pre-soaked overnight and tied with gunny bags and allowed to germinate during

109

two days. Sprouted seeds were directly planted in pots and excess seedlings were thinned out, one week

110

after planting, to achieve uniform density. Fertilizer management and the other crop management

111

practices were followed according to the recommendations of Department of Agriculture [13].

112 113

Five cultivated rice varieties (Bg300, Bg352, At362, Bg379-2 and H4) were used to determine the

114

suitable glyphosate concentration at the preliminary level. Five different concentrations of glyphosate

115

(0.25 g/l, 0.5 g/l, 1 g/l, 2 g/l and 3 g/l) were applied using hand sprayer at three different age levels (2, 3,

116

and 4 weeks after sowing-WAS). Comparatively low concentrations of glyphosate were used in the

117

experiment because higher concentration precludes the adaptive progress of the rice plant [14].

118

Subsequently, the suitable concentration of glyphosate determined was applied to 25 cultivated rice

119

varieties to evaluate glyphosate resistance. Complete Randomize Design (CRD) was followed with three

120

replicates and replicates with non- glyphosate treatment served as the control.

121

application of glyphosate, the dead plants were considered as susceptible to the herbicide and surviving

122

plants with a substantial growth as resistant to the herbicide. For each rice variety, the number of resistant

123

plants and total number of plants were counted. The percentage (%) of resistance was calculated using

124

the equation (1). Plants with ≥ 50% resistance to glyphosate (0.5 g/l) were arbitrary considered as

125

resistant varieties.

Subsequent to the

4

126 127

(1)

128 129

Method 2

130

Seeds from each variety were pre-soaked in distilled water for 24 h at room temperature. The pre-soaked

131

seeds were separated into two portions and one was served as the control. Based on the results of the

132

previous studies, the other portion of seeds was exposed to 4.5 mmol/l EMS for 12 hours. The seeds were

133

then washed with running tap water and allowed to leach the residual chemicals and placed in Petri

134

dishes containing of moist filter papers. These Petri dishes were kept in the moisture chamber. The

135

germinated seedlings were transferred into mud pots and glyphosate was applied at 2WAS (S0 -First

136

generation).

137

Following the application of glyphosate, the dead plants were considered as susceptible to the herbicide

138

and surviving plants with a substantial growth were considered as resistant to herbicide. For each rice

139

variety, the number of resistant plants and percentage resistance was calculated. Plants with ≥ 50%

140

resistance to glyphosate (0.5 g/l) were arbitrary considered as resistant varieties.

141

The morphological and yield data were collected from the resistant varieties and seeds of self-pollinated

142

S0 generation of mutated plants were designated as S1 seeds. Three panicles per survived S1 (Second

143

generation) plants were harvested from mutated resistant plants and evaluate their resistance to

144

glyphosate using the same procedure. The experiment was repeated for three times of cropping seasons

145

of the country.

146 147

Statistical analysis

148 149

Descriptive statistics were performed on the dataset. The mean and Standard deviation was computed and

150

ANOVA tests were used to compare the mean. One-way-analysis of variance (ANOVA) was performed

151

on agro-morphological characters. All statistical analyses were carried out using SAS Version 9.2 (SAS,

152

2008).

153 154

The data were furthermore subjected to neural network analysis for developing models other than the

155

descriptive analysis. The artificial neural networks (ANN) have widely been used across different fields

156

of studies in complex problems involving nonlinearity and uncertainty. With advancement of computer

157

technology, computation power and new algorithms have emerged and increase the application of neural 5

158

networks in many field of studies in Soil microbiology [14],

159

including agricultural sciences resulted. The Self-Organizing Map (SOM) is one of the unsupervised

160

types of ANN. SOM projects the multi-dimensional data into a two dimensional space [16]. The SOM

161

includes two layers i.e., input and output layer. The input layer serves as receiver of input variables and

162

the neurons in the output layer presents patterns reflected from the input data in two dimensional space in

163

a form that facilitate the better visualization of the results.

164 165

The SOM model developed using the SOM toolbox in Matlab (2010). Initially, a number of SOM

166

models were developed with varying number of output neurons for clear grouping patterns of treatments.

167

Based on the results of the initial study, a SOM model was chosen which include in the present study

168

consisted

169 170

During the training the SOM model default values of learning parameters and map topology was used.

171

Subsequently, the different selection test measures were calculated to evaluate the quality of the map the

172

best SOM. Model performance was measured by the quantization error which indicates the average

173

distance between each data vector and associated best matching unit (BMU). The quantization error (Qe)

174

is calculated as shown in equation (2), where N is the number of data-vectors, Xi is the current vector

175

patterns, mc is the best matching neuron (BMN) of the corresponding Xi input vector.

176 177

(2)

178 179

The quantization error was use to evaluate the goodness of fitting the clustering of treatment in the map

180

to the data. Thus, the optimal map produces the smallest average quantization error. The smaller the

181

quantization error, the smaller the average of the distance from the vector data to the prototypes, i.e. the

182

data vectors is closer to its prototypes.

183 184

As an alternative method in measuring fitness of a SOM, topographic error (Et) was calculated [17] [18].

185

The Et, was calculated according to equation (3) where U (xi) = 1, if the 1st BMN and 2nd BMN are not

186

adjacent, otherwise U (xi) = 0. The lower topographic error (Et) represent the better preservation of SOM

187

topology.

188 189 6

190

(3)

191 192

The best SOM was selected based on the error performance of Quantization Error (

193

Topographical Error (

) and

).

194 195 196 197

RESULTS

198 199

The results of the preliminary study revealed that 0.5 g/l at 3WAS was the most suitable concentration at

200

which most of the rice varieties showed glyphosate resistance. At 0.25g/l glyphosate concentration,

201

plants of all selected varieties were survived and with the increasing concentrations (1 g/l, 2g/l and 3 g/l)

202

there were reduction in herbicide resistance (Fig. 1).

203 204

The magnitude of glyphosate resistant values indicates that Bg300, Bg304, H4 and “Rathhal” were

205

susceptible to (<35%) 0.5g/l glyphosate concentration and Bg94-1, Bg250, Bg304, Bg352, Bg357,

206

Bg360, Bg450, “Suwadel” and “Madel” showed slight resistance (<50%) (Fig.3). Meanwhile, twelve

207

cultivated rice varieties (Bg359, At362, Bw364, Ld365, Bg366, Bg369, Bg379-2, Bg403, Bg454,

208

"Pachcha perumal”, “Kalu heenati” and “Kurulu thuda”) indicated a considerably higher resistance

209

(≥50%) compared to other varieties at 0.5g/l glyphosate concentration.(Fig.2).

210

Fig 1: Resistant percentage of rice at different glyphosate concentrations and different application times

211

Fig 2: Percentage resistant of different rice varieties at 0.5 g/l glyphosate concentration

212

The analysis of variance of the agro-morphological characters and yield parameters clearly shows that

213

there were statistically significant differences between control plants and glyphosate treated plants.

214

However, on visual observation, it was found that plant height was slightly retarded in the treated plants

215

compared to control plants. Meanwhile, there was no noticeable difference in the number of leaves and

216

number of tillers with the application of glyphosate. All the yield parameters (number of panicle /plant,

217

number of seeds/panicle and 1000 grain weight) indicated significant difference compare to controls.

218

However, there were also some visual injuries with chlorosis in the upper part of the leaves but newly

219

emerged leaves were remaining green in color. Multiple shoots were aroused from the internodes of main 7

220

stem (Fig.3: a) and the secondary shoots (“Kalu heenati” and “Pachcha perumal”) and flag leaves were

221

wrinkled or curled. At the booting stage all the leaves of “Kurulu thuda” variety were curled and leaf

222

discoloration occurred. Until the harvesting stage plants exist in the same stage and no longer show any

223

maturity (Fig.3: b).

224 225

Figure 3: Visual injuries caused by glyphosate, a) Multiple shoots and roots aroused from the internodes,

226

b) Leaf curling and discoloration, c) Fused panicle to flag leaf, d) Bleached lemma and palea

227

Future more, malformation of inflorescences also observed in some varieties throughout reproductive

228

stage. The florescence of Bg369 and Bg454 were got aborted inside the flag leaf sheath and unable to

229

emerge as a panicle (Fig.3: c). some panicles were unable to fully appear due to the fusing to the flag leaf

230

in the maturity stage. Inflorescence malformation and developing grains with only bleached lemma and

231

palea were common in Bg366 (Fig.3: d). In Bg379-2 variety distorted and crescent-shape spikelet were

232

observed and it seemed to be associated with the “Parrot beaked” one of the straightforward

233

physiological disorder of rice. [19].

234

EMS Study

235

Results indicated that fourteen varieties increased the resistance to glyphosate (Bg94-1, Bg352, Bg359,

236

Bg360, At362, Bw364, Ld365, Bg366, Bg379-2, Bg403, Bg454, “Kalu heenati”, “Pachcha perumal” and

237

“Madel”) in the first generation -S0 (Fig.4).

238

Figure 4: Percentage resistant of different rice varieties at 0.5 g/l glyphosate concentration after EMS

239

mutation

240

The pair-wise statistical analysis of variance of agro-morphological characters clearly showed significant

241

differences (p ≤ 0.05) between control and treated plants. The percentage of resistance in S1 was almost

242

similar in all the varieties except Bg454 and “Madel”. In contrary, there were statistically significant

243

differences (p ≤ 0.05) between the EMS-mediated-mutated rice plants with non-mutated rice plants

244

related to agro-morphological and yield characters such as plant height, number of panicles/plant,

245

Seeds/panicle and 1000 grain weight. However, number of leaves/plant and number of tillers /plant

246

statistically insignificance when compared mutated rice plants with non-mutated rice plants.

247

Table 1: Variation of Agro-morphological and yield characters of rice varieties (Natural screening, EMS

248

treated). The mean value is followed by standard error within parenthesis. 8

249

Figure 5: Umatrix and the label matrix of the SOM developed from the agro-morphological and yield

250

characters (C: control for natural screening, G: natural screened plants/ treated with glyphosate, E: S0

251

generation mutated, CS: control in S1 generation, ES: S1 generation mutated)

252

The Umatrix resulted from SOM is shown in Fig.6 indicates that there is a discernible grouping tendency

253

which reflects the effect of treatments viz. glyphosate and EMS (two generations). The careful

254

observation of components planes of the SOM revealed that certain agro-morphological and yield

255

characters considerably vary across the treatment. In particular, number of leaves per plant; filled grain

256

percentage; 1000 grain weight was considerably varied across the treatments. Meanwhile, the variation of

257

agro-morphological characters such as plant height (cm), number of tillers /plant, number of

258

panicles/plants and seeds per panicle were less and indicate that there was no considerable penalty with

259

respect to the treatments.

260

In careful observation of component planes (Fig.6) revealed that controls and glyphosate treated four

261

varieties such as BW364, Ld365, Bg369 and “Kalu heenati” grouped to a same cluster indicating

262

resistance to glyphosate. More or less similar patterns were observed in EMS mutated rice varieties

263

(Bg359, At362, Bg403 and “Kalu heenati”)

264

Figure 6: SOM planes of agro-morphological and yield characters used in the study. (PH: Plant height,

265

Till: Number of tillers/plant, LV: Number of leaves /plant, PAN: Number of panicles/plant, SEE:

266

Number of seeds/panicle, WEI: 1000 grain weight, PER: Filled grain percent.

267 268 269 270 271 272 273

DISCUSSION

274 275

The results of the field trial revealed that the response of rice varieties to glyphosate was varied according

276

to the concentration and the age of application. The similar trends have been reported by [20] Ellis et al.

277

2013 with the application of glyphosate. After the application of glyphosate there was a general chlorosis

278

in the uppermost levels and then the plants died. But when applying the same concentration of glyphosate

9

279

at three WAS, it showed a diversity of response and 3WAS was the most suitable age for the selecting of

280

HR rice varieties.

281

Although 25 rice varieties used for the study, there were only twelve varieties showing a considerable

282

resistance against glyphosate (Bg359, At362, Bw364, Ld365, Bg366, Bg369, Bg379-2, Bg403, Bg454

283

“Pachcha perumal”, “Kalu heenati” and “Kurulu thuda”) at seedling stage. Even though a formula has

284

been used to calculate the resistant percentage, visual observations of the treated plants also considered

285

throughout the growth cycle. Glyphosate had a greater impact on Bg300, Bg304, H4 and “Rathhal” rice

286

varieties. Bg94-1, Bg250, Bg304, Bg352, Bg357, Bg358, and Bg360 and “Suwadal” were showing slight

287

resistant (35≥50) to glyphosate, but most of the plants had the visual injuries after exposure to herbicide.

288

The newest leaf to emerge following treatment often emerged tightly rolled and the overall plant stunting

289

was observed. During the vegetative growth phase plants showed various forms of malformations of

290

leaves and tillers. Most of the varieties categorized under slight resistant were not able to recover the

291

malformation during the vegetative stage, hence not producing panicle except by 352 and “Suwadel”.

292

Bg352 showed insignificant recovery in vegetative phase and heading of the panicles were delayed thus

293

produced empty kernels. The traditional variety “Suwadel” was recovered within two to three weeks after

294

treatment (WAT) and formed some productive tillers.

295 296

Almost all the varieties categorized as resistant to glyphosate (Bg359, At362, Bw364, Ld365, Bg366,

297

Bg369, Bg379-2, Bg403, Bg454, Pachcha perumal”, “Kalu heenati” and “Kurulu thuda”) showed

298

recovery within two to three WAT. Morphological data such as plant height was negatively affected by

299

the application of glyphosate which leads to overall plant stunting compare to control plants. [21] also

300

reported the similar results after application of same herbicide to rice. The severe visual injuries plants

301

were observed during vegetative phase compared to the reproductive stage which may be due to the

302

translocation of glyphosate into meristematic tissues [22].

303 304

Comparison of yield parameters of natural resistant rice varieties such as At362, Bw364, Ld365 and

305

Bg403 revealed that maximum number of panicles per plant and the number of seeds per panicle were

306

higher than other varieties included in the study. The EMS treated plant possibly generates mutation with

307

both positive and negative traits. However, the above mentioned four rice verities indicated enhanced

308

resistance to glyphosate after the mutagenesis.

309

10

310

Mutated and non-mutated plants of above-mentioned rice varieties indicated higher recovery rate and

311

significant growth throughout the vegetative growth phase. The similar findings [23] also reported that

312

certain rice cultivars can overcome injury better than others. The results obtained in this study supports

313

the conclusions made by the previous studies revealed that some of the traditional and developed

314

cultivated rice varieties in Sri Lanka possess a natural resistance to glyphosate [24].

315 316

In addition, the study showed that EMS could successfully be used to induce HR in glyphosate

317

susceptible rice varieties by excluding negative mutations and to enhance the existing HR. The chemical

318

mutagen EMS has been used to develop HR varieties against the broad-spectrum herbicide,

319

Imidazolinone [25; 26]. The present results revealed that an increase in existing natural HR in most of the

320

varieties subsequent to the EMS mutagenesis process. The comparison of resistant percentage induced

321

and enhanced in S0 and S1 equally and which imply the resistance has transferred and stable. Though the

322

EMS mutant plants showed a slight variation in reduction of agro-morphological characters, the HR

323

induced varieties can be used in rice breeding programs for the development of HR rice varieties. The

324

application of SOM in HR studies is rewarding and can be used to visualize the results which lead to easy

325

interpretation. Further SOM can be used as a complementary analysis of the results abstained from the

326

conventional statistical procedure.

327 328 329

CONCLUSIONS

330 331

In developing HR rice varieties, it is crucial to understand the most effective glyphosate concentration.

332

An effective concentration will lead to evaluate the existence of natural HR in rice varieties. Developing

333

broad-spectrum HR rice varieties via classical breeding methods such as mutation with EMS provides

334

novel efficient chemical weed controlling method in rice cultivation. Meanwhile, HR rice will also have

335

led to reduce the amount of herbicide and frequency of application. Therefore, naturally existing HR rice

336

varieties have a higher potential in booster the herbicide resistance through mutation via EMS in rice

337

breeding programs leading to develop new HR rice varieties in future. The complexity of the data

338

preclude the visualization of the patterns emerged and the Self Organizing Map (SOM) is a rewarding

339

tool to visualize the higher dimensional data in a low dimensional space highlight the patterns reflected

340

from the data related to agricultural researches. Future extensive work is required for number of

341

generation of the EMS mutated rice genotype to establish herbicide resistant lines. In addition, molecular 11

342

analysis (AFLP) is being carried out for confirmation of HR varieties and identification of potential

343

AFLP markers for HR.

344 345

ACKNOWLEDGEMENTS

346

The research grant provided by the National Research Council (NRC), Sri Lanka (Grant No. NRC 12-

347

037) is greatly appreciated and proper guidance of Mrs. A.S.K. Abeysekara, Senior Research Officer,

348

RRDI, Batalagoda, Sri Lanka is also greatly acknowledged.

349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383

REFERANCES [1] Rabbani NR, Bajwa JA. Interference of five problematic weed species with yield.Afri. J. Biotech .2011 ; 10(10) :1854- 1862

rice growth and

[2] Abeysekara ASK. Management of Echinochloa spp. in rice in Sri Lanka. FAO workshop on Echinochloa spp. Control, Beijing, China.2011. p.13 [3] Tu M, Hurd C, Randall JM. Weed control methods handbook: tools & techniques for use in natural areas, The Nature Conservancy, Wild land Invasive Species Team, version April 2001. [4] Juraimi AS, Uddin K, Anwar P, Mohamed MTM, Ismail R ,Man A . Sustainable weed management in direct seeded rice culture: A review. AJCS 2013; 7(7):989-1002 [5] Senseman SA .Herbicide Handbook. 9 th ed. Lawrence, KS: weed Science Society of America; 2007.p.458. [6] Labrada R. Weedy and wild rices: Their impact and management. Procceding of the 21st Asian Pacific Weed Science. Colombo,Sri Lanka; 2007.p.8-15. [7] Davis BM, Scott RC, Norsworthy JK, Smith KL. Effects of low rates of glyphosate and glufosinate on rice. Arkansas Agricultural Experiment Station Research Series 2009; pp: 581. [8] Technical note, Weed Technology. 1998; 12 (4):789

[9] Cioppa G, Bauer SC, Klein BK, Shah DM, Fraley RT, Kishore GM. Translocation of the precursor of 5-enolpyruvylshikimate-3-phosphate synthase into chloroplasts of higher plants in vitro. Proceedings of the National Academy of Sciences. 1986 Sep 1; 83(18):6873-7. [10] Salas RA, Dayan FE, Pan Z, Watson SB, Dickson JW, Scott RC, Burgos NR. EPSPS gene amplification in glyphosate‐resistant Italian ryegrass (Lolium perenne ssp. multiflorum) from Arkansas. Pest management science. 2012; 68(9):1223-30.

12

384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405

[11] Guttieri MJ, Eberlein CA,Smith DCM. Molecular genetics of target-site resistance to acetolactate synthase inhibiting herbicides.1999; p. 645.

406

[18] Kohonen T. Software Tools for SOM. Self-organizing maps, 2001; 30 (3): 311-328.

407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429

[12] Beckie HJ, Harker KN, Hall LM, Warwick SI, Légère A, Sikkema PH, Clayton GW, Thomas AG, Leeson JY, Séguin-Swartz G, Simard MJ. A decade of herbicide-resistant crops in Canada. Canadian Journal of Plant Science. 2006; 86(4):1243-64. [13] Department of Agriculture. Crop recommendation. Link: http://www.doa.gov.lk/index.php/en/croprecommendations/808/2013 [14] Young IM, Crawford JW. Interactions and self-organization in the soil-microbe complex. Science. 2004; 304(5677):1634-7. [15] Brosse S, Giraudel JL, Lek S. Utilisation of non-supervised neural networks and principal component analysis to study fish assemblages. Ecological Modelling. 2001; 146(1):159-66. [16] Kohonen JT .Self-Organizing Maps. Vol. 30.of Springer Series in Information Sciences, 3rd ed., 2001; Springer-Vela, Berlin Heidelberg.

[17] Kohonen T. Self-organized formation of topologically correct feature maps. Biol. Cybern. 1982; 43, 59–69.

[19] Groth D, Hollier C, and Rush C. Disease management. In: J.Saichuk,editors. Louisiana Rice Production Handbook. Baton Rouge, LA: Lousiana State University Ag Center 2009.p .72-92.

[20] Ellis JM, Griffin JL, Linscombe SD, Webster EP. Rice (Oryza sativa) and Corn (Zea mays) Response to Simulated Drift of Glyphosate and Glufosinate 1. Weed Technology. 2003 ; ( 3):452-60. [21] Koger CH, Shaner DL, Krutz LJ, Walker TW, Buehring N, Henry WB, Thomas WE, Wilcut JW. Rice (Oryza sativa) response to drift rates of glyphosate. Pest management science. 2005; 61(12):1161-7. [22] Martin RA, Edgington LV. Comparative systemic translocation of several xenobiotics and sucrose. Pesticide Biochemistry and Physiology. 1981; 16(2):87-96. [23] Meier JR, Smith KL, Scott RC, Doherty RC. Rice Cultivar response to Low Glyphosate Rates as Influenced by Growth Stage. AAES research series 2006; 550:205-210. [24] Weerakoon SR, Somaratne S, Wijerathne RDG, Ekanayaka EMSI. Natural Herbicide Resistance (HR) to Broad –spectrum Herbicide, Glyposate among Traditional and Inbred – cultivated rice (Oryza sativa L) Varieties in Sri Lanka. Pakistan journal of Biological Sciences 2013; 16 (16):796-803. 13

430 431 432 433 434 435 436

[25] Gealy DR, Mitten DH, Rutger JN. Gene flow between red rice (Oryza sativa) and herbicide-resistant rice (O. sativa): implications for weed management 1. Weed technology. 2003; 17(3):627-45. [26] Lang NT, Buu BC. Rice breeding and inheritance of herbicide resistance in Clearfield rice (Oryza sativa L.).Omonrice 2007; 15: 36-45.

437

Table 1: Variation of Agro-morphological and yield characters of rice varieties (Natural screening, EMS

438

treated). The mean value is followed by the standard error within parenthesis. Rice

Code

Plant. Height

Variety

No. of

No.

.till

of

No. of.

Seed/

leaves

panicle

Panicle

Filled grain

1000 grain

Bg359C

1C

62.00(2.08)

2

14

7

83

75.37(0.32)

22.46(0.64)

Bg359T

1G

52.00(3.06)

2

8

5

61

69.60(0.29)

16.37(0.38)

At362C

2C

66.33(1.20)

2

8

5

188

84.57(0.59)

25.00(0.44)

At362T

2G

50.33(0.88)

1

5

5

155

82.87(0.24)

16.86(0.32)

Bw364C

3C

62.33(1.45)

2

5

6

132

85.17(0.49)

24.21(0.42)

Bw364T

3G

55.00(3.40)

2

4

4

113

70.33(2.97)

14.26(0.55)

Ld365C

4C

59.00(2.08)

2

12

4

75

81.27(0.50)

13.26(0.29)

Ld365T

4G

48.00(2.00)

3

9

3

57

74.97(2.98)

9.85(0.20)

Bg366C

5C

46.67(2.73)

2

9

4

80

72.27(0.44)

23.17(0.27)

Bg366T

5G

27.67(3.67)

2

6

2

67

64.27(0.12)

18.49(0.90)

Bg369C

6C

34.33(1.67)

2

5

Bg369T

6G

48.67(2.33)

2

8

Bg379C

7C

64.00(0.58)

3

9

3

61

68.70(0.28)

25.67(0.44)

Bg379 T

7G

48.00(6.00)

2

5

2

54

66.30(0.35)

17.10(1.85)

Bg403C

8C

67.00(1.15)

3

12

5

142

82.43(0.67)

21.39(0.38)

Bg403T

8G

51.00(3.00)

2

12

4

108

80.95(0.45)

17.79(0.54)

Bg454C

9C

52.00(2.08)

2

10

Bg454T

9G

45.33(1.45)

2

7

KTC

10C

70.00(1.15)

2

10

KTT

10G

64.33(2.33)

1

6

KHC

11C

73.33(0.88)

2

10

3

69

70.93(0.54)

22.89(1.51)

KHT

11G

59.83(3.68)

1

8

4

58

68.50(0.36)

14.37(0.59)

PPC

12C

70.33(1.20)

2

11

5

77

80.23(1.13)

31.64(0.38)

PPT

12G

61.00(3.12)

1

8

3

49

70.63(0.30)

22.94(1.34)

Bg94C

13C

88.50(0.29)

4

12

3

70

76.80(1.14)

18.80(0.38)

Bg94T

13E

78.67(5.04)

3

8

3

61

69.23(6.65)

17.43(0.35)

14

Bg352C

14C

88.00(0.58)

3

14

3

89

70.37(1.29)

20.89(0.44)

Bg352T

14E

78.67(1.86)

2

13

2

87

66.40(1.95)

18.03(0.09)

Bg359C

15C

99.00(0.58)

4

13

3

77

71.37(0.44)

21.33(0.50)

Bg359T

15E

89.00(0.88)

3

12

2

50

68.13(0.67)

18.23(0.58)

Bg360C

16C

97.33(1.20)

3

12

3

68

77.90(1.53)

12.63(0.09)

Bg360T

16E

78.67(2.33)

2

10

2

58

67.13(1.07)

11.30(0.29)

At362C

17C

102.67(3.71)

3

12

4

131

83.87(1.74)

22.93(0.83)

At362T

17E

92.33(1.45)

3

9

3

112

82.17(1.16)

18.87(0.20)

Bw364C

18C

105.00(2.88)

3

10

4

160

87.13(0.92)

22.60(0.67)

Bw364T

18E

85.33(2.03)

3

9

2

95

69.23(2.99)

18.20(0.71)

Ld365C

19C

101.33(1.86)

3

16

4

117

81.60(0.72)

13.33(0.41)

Ld365T

19E

72.67(1.45)

3

12

3

104

71.23(1.01)

10.05(0.10)

Bg366C

20C

93.00(0.58)

3

17

4

155

71.33(0.66)

20.83(0.24)

Bg366T

20E

85.00(0.58)

3

15

3

85

61.43(0.69)

18.27(0.46)

Bg369C

21C

109.00(2.08)

3

15

3

132

68.17(1.17)

21.23(0.73)

Bg369T

21E

82.00(2.31)

2

11

Bg379C

22C

88.33(2.85)

3

14

3

88

68.27(1.16)

21.50(0.32)

Bg379 T

22E

72.67(1.45)

3

13

2

107

62.97(1.35)

17.63(0.43)

Bg403C

23C

102.33(3.84)

3

17

4

108

82.03(0.26)

21.87(0.24)

Bg403T

23E

94.33(1.20)

3

11

3

115

79.73(0.79)

18.80(0.32)

Bg454T

24E

76.00(1.53)

3

12

Bg454C

24C

90.33(0.88)

3

14

3

100

78.97(0.63)

17.47(0.40)

PPC

25C

162.33(1.45)

4

16

4

101

78.07(1.21)

21.57(0.58)

PPT

25E

150.67(1.76)

3

11

3

91

68.33(0.50)

19.30(0.57)

KHC

26C

111.67(6.01)

3

14

4

84

68.17(1.59)

21.43(0.81)

KHT

26E

115.00(2.89)

3

11

3

93

65.63(1.80)

19.40(0.47)

MDC

27C

154.00(2.08)

5

15

3

82

19.97(0.30)

70.43(1.38)

MDT

27E

141.00(2.08)

3

12

2

68

18.27(0.50)

64.73(0.93)

Bg94T

28CS1

86.33(0.67)

3

12

3

69

75.47(20.5)

18.27(0.15)

Bg94C

28ES1

75.67(1.20

2

9

2

53

74.27(5.45)

16.43(0.35)

Bg352C

29CS1

82.67(1.20)

2

14

2

84

68.93(1.21)

18.70(0.31)

Bg352T

29ES1

69.67(4.26)

1

11

2

81

63.63(0.81)

15.17(-0.26)

Bg359C

30CS1

95.33(0.88)

4

11

3

78

70.27(0.58)

20.10(0.67)

Bg359T

30ES1

82.37(1.43)

3

11

3

43

65.90(0.60)

17.40(0.32)

Bg360C

31CS1

82.40(6.83)

3

10

3

64

74.97(0.77)

11.17(0.20)

Bg360T

31ES1

72.00(0.58)

3

8

2

54

61.67(1.07)

10.03(0.12)

At362C

32CS1

97.33(1.20)

2

12

4

131

83.87(1.74)

22.93(0.83)

15

At362T

32ES1

87.67(1.45)

2

9

3

112

82.17(1.16)

18.87(0.20)

Bw364C

33CS1

105.00(2.89)

3

10

4

160

87.13(0.92)

22.60(0.67)

Bw364T

33ES1

85.33(2.03)

3

9

2

95

69.23(2.99)

18.20(0.17)

Ld365C

34CS1

97.67(1.45)

2

15

4

117

81.60(0.72)

13.33(0.19)

Ld365T

34ES1

70.00(1.15)

2

10

3

104

71.23(0.66)

10.05(0.10)

Bg366C

35CS1

91.00(0.58)

2

15

4

155

71.33(0.66)

20.83(0.24)

Bg366T

35ES1

81.67(0.33)

2

11

3

85

61.43(0.69)

18.27(0.46)

Bg379C

36CS1

84.67(2.60)

2

13

2

90

80.53(0.35)

19.90(0.10)

Bg379 T

36ES1

75.00(1.45)

2

8

1

76

75.57(1.48)

14.77(1.51)

Bg 403 C

37CS1

84.67(2.06)

2

13

2

90

80.53(0.35)

19.90(0.10)

Bg 403 T

37ES1

75.00(1.73)

2

8

1

76

75.57(2.00)

14.77(0.58)

PPC

38CS1

151.33(1.20)

4

16

3

94

75.23(2.20)

20.60(0.45)

PPT

38ES1

119.00(3.51)

2

10

2

85

60.37(0.70)

14.43(0.41)

KHC

39CS1

102.33(1.45)

2

13

3

80

59.10(3.94)

20.00(0.89)

KHT

39ES1

89.00(2.08)

2

9

2

79

59.53(4.79)

19.40(0.47)

439 440

16

441 442

17

443 444

18

445 446

19

447 448

20

449 450

21

451

22