Download - 1 Short title: Semi-automated Sorghum Panicle Phenotyping 2 · 157 panicle phenotypes from a subset of the Sorghum Association Panel (SAP) (Casa et al., 158 2008) consisting of 272

1

Short title Semi-automated Sorghum Panicle Phenotyping 1

2

3 Corresponding Author 4 5 Patrick S Schnable 6 Department of Agronomy 7 Iowa State University 8 2035B Roy J Carver Co-Lab 9 Ames IA 50011 10 Phone +1(515)294-0975 11 Email schnableiastateedu 12 Research Area Breakthrough Technologies 13

14

15

16

17

18

19

20

21

22

23

24

Plant Physiology Preview Published on November 2 2018 as DOI101104pp1800974

Copyright 2018 by the American Society of Plant Biologists

wwwplantphysiolorgon March 3 2020 - Published by Downloaded from Copyright copy 2018 American Society of Plant Biologists All rights reserved

2

Semi-automatedfeatureextractionfromRGBimagesforsorghumpanicle25 architectureGWAS26 Yan Zhou1 Srikant Srinivasan12 Seyed Vahid Mirnezami3 Aaron Kusmec1 Qi Fu45 Lakshmi 27 Attigala1 Maria G Salas Fernandez1 Baskar Ganapathysubramanian3 Patrick S Schnable14 28

1Department of Agronomy Iowa State University Ames Iowa 50011 USA 29

2Current Address School of Computing and Electrical Engineering Indian Institute of 30 Technology Mandi Kamand Himachal Pradesh 175005 India 31

3Department of Mechanical Engineering Iowa State University Ames 50011 Iowa USA 32

4College of Agronomy China Agricultural University 100083 Beijing China 33

5Current Address Molecular Breeding Lab BGI Research 518083 Shenzhen China 34 35

One-sentencesummary Image-based phenotyping facilitates the genetic dissection of 36 uni- and multi-dimensional traits of sorghum panicles 37

AuthorContributions38 PSS and BG conceived the original research plan and supervised the experiments and 39 data analyses PSS YZ and SS designed the experiments SS and SVM developed the 40 phenotype extraction pipeline and conducted phenotype extractions YZ conducted most 41 of the experiments and analyses QF assisted with collecting phenotypic data and image 42 processing AK assisted with GWAS and genetic analyses MGSF assisted with the 43 interpretation of results PSS LA MGSF and YZ wrote the article with contributions 44 from all the authors 45

Fundingsources46 This material is based upon work that was supported in part by the National Institute of 47 Food and Agriculture US Department of Agriculture (award no 2012-67009-19713) to 48 PSS and MGSF and colleagues and additional funding from ISUrsquos Plant Sciences Institute 49 to PSS MGSF and BG 50 Address correspondence to schnableiastateedu 51 52 53

Abstract 54


3

Because structural variation in the inflorescence architecture of cereal crops can influence 55 yield it is of interest to identify the genes responsible for this variation However the 56 manual collection of inflorescence phenotypes can be time-consuming for the large 57 populations needed to conduct GWAS (genome-wide association studies) and is difficult for 58 multi-dimensional traits such as volume A semi-automated phenotyping pipeline (Toolkit 59 for Inflorescence Measurement TIM) was developed and used to extract uni- and multi-60 dimensional features from images of 1064 sorghum (Sorghumbicolor) panicles from 272 61 genotypes comprising a subset of the Sorghum Association Panel (SAP) GWAS detected 35 62 unique SNPs associated with variation in inflorescence architecture The accuracy of the 63 TIM pipeline is supported by the fact that several of these trait-associated SNPs (TASs) are 64 located within chromosomal regions associated with similar traits in previously published 65 QTL and GWAS analysis of sorghum Additionally sorghum homologs of maize (Zea mays) 66 and rice (Oryza sativa) genes known to affect inflorescence architecture are enriched in the 67 vicinities of TASs Finally our TASs are enriched within genomic regions that exhibit high 68 levels of divergence between converted tropical lines and cultivars consistent with the 69 hypothesis that these chromosomal intervals were targets of selection during modern 70 breeding 71

Introduction72 The grass family (Poaceae) includes maize (Zea mays) wheat (Triticum aestivum) rice 73 (Oryza sativa) sorghum (Sorghum bicolor) and other cereal crops which collectively 74 provide 56 of the calories consumed by humans in developing countries and over 30 in 75 developed countries (Amine et al 2003 Bruinsma 2017) The development of the grain-76 bearing inflorescences of cereals begins with the transition of the vegetative shoot apical 77 meristem (SAM) into an inflorescence meristem which later forms into branch meristems 78 and further generates spikelet meristems (Zhang and Yuan 2014) Variation in these 79 developmental processes accounts for the substantial interspecific and intraspecific 80 variability in inflorescence architecture observed among the cereals (Vollbrecht et al 2005 81 Huang et al 2009 Youssef et al 2016) The transition of the SAM into an inflorescence 82 meristem is regulated by genes that affect both the identity and maintenance of meristems 83 (Pautler et al 2013 Zhang and Yuan 2014) For example in Arabidopsis (Arabidopsis84 thaliana) meristem identity is primarily regulated by a negative feedback loop between 85 CLAVATA (CLV) and the homeobox gene WUSCHEL (WUS) which prevents the 86 misspecification of meristem cells and premature termination of floral and shoot 87 meristems (Laux et al 1996 Mayer et al 1998 Pautler et al 2013 Tanaka et al 2013) 88 Mutations of CLV genes often result in larger inflorescence meristems (Clark et al 1997 89 Fletcher et al 1999 Jeong et al 1999) Similarly mutations of the CLV1 homolog in maize 90 thicktasseldwarf1 (TD1) and the CLV2 homolog fasciatedear2 (Fea2) produce tassels 91 with more spikelets and fasciated ears with extra rows of kernels (Taguchi-Shiobara et al 92


4

2001 Bommert et al 2005) The KNOTTED1-like homeobox (KNOX) genes also affect 93 inflorescence development by altering the establishment and maintenance of SAM tissues 94 (Vollbrecht et al 2000 Bolduc and Hake 2009 Bolduc et al 2012) Mutations of the maize 95 knotted1 (Kn1) and the rice Oryzasativahomeobox1 (OSH1) genes both exhibit a ldquosparse 96 inflorescencerdquo phenotype caused by reduced meristem maintenance (Kerstetter et al 97 1997 Tsuda et al 2011) The null allele kn1-E1 is epistatic to the null allele td1-glf in maize 98 ear development and suggests the importance of Kn1 in regulating both meristem identity 99 and lateral organ initiation (Lunde and Hake 2009) 100 These functional studies of large-effect or qualitative mutants have greatly enhanced our 101 understanding of the developmental processes underlying inflorescence development and 102 architecture Even so because many quantitative traits are also affected by large numbers 103 of small-effect genes (Buckler et al 2009 Danilevskaya et al 2010 Brown et al 2011 Li 104 et al 2012) there remains the opportunity to expand our understanding of inflorescence 105 development via the application of GWAS (genome-wide association study) to identify 106 associations between specific loci and quantitative phenotypic variation GWAS has been 107 used to identify genes associated with inflorescence architecture in multiple crops (Brown 108 et al 2011 Morris et al 2013 Crowell et al 2016 Wu et al 2016 Zhao et al 2016 Xu et 109 al 2017) Given that thousands or millions of markers can now be readily discovered and 110 genotyped phenotyping is typically the bottleneck for conducting GWAS Traditionally crop 111 scientists have collected unidimensional traits such as spike length spike width and 112 branch length and number manually This is time-consuming for large populations 113 Therefore to fully utilize the advantages of GWAS there is a need for accurate high-114 throughput phenotyping platforms 115 Computer vision has been shown to be efficient in isolating inflorescences (eg Aquino et 116 al 2015 Zhao et al 2015 Millan et al 2017) and several studies have attempted to extract 117 inflorescence features from images of rice and maize (AL-Tam et al 2013 Crowell et al 118 2014 Zhao et al 2015 Gage et al 2017) The complexity of inflorescence architecture 119 complicates the accurate extraction of phenotypes from images To date two studies have 120 applied image-based phenotyping to the genetic analyses of crop inflorescences and they 121 either focused only on artificially flattened rice inflorescences (Crowell et al 2016) or on 122 unidimensional traits of maize such as tassel length and central spike length (Gage et al 123 2018) Considering that inflorescences are three-dimensional (3D) structures phenotyping 124 strategies that flatten an inflorescence or focus on only a single plane will inevitably fail to 125 capture a considerable amount of phenotypic variation and will therefore reduce the 126 probability of discovering genes involved in inflorescence architecture This limitation 127 highlights the need for new automated phenotyping platforms that accurately collect 128 inflorescence traits especially multi-dimensional traits which have not been collected by 129 previous automated phenotyping projects 130


5

Sorghum is the worldrsquos fifth most important cereal crop and is a major food crop in some 131 developing countries (Hariprasanna and Rakshit 2016) It is evolutionarily closely related 132 with the other well studied cereals such as maize wheat and rice (Paterson et al 2009 133 Schnable et al 2012 Choulet et al 2014 Schnable 2015) Studies conducted to date on 134 the genetic architecture of sorghum panicles have focused on unidimensional traits such as 135 panicle length panicle width and branch length (Hart et al 2001 Brown et al 2006 136 Srinivas et al 2009 Morris et al 2013 Nagaraja Reddy et al 2013 Zhang et al 2015 137 Zhao et al 2016) Due to challenges associated with collecting multi-dimensional traits 138 such as panicle area volume and compactness our understanding of the genetic 139 architectures of these traits is limited 140 Sorghum exhibits extensive population structure associated with both morphological type 141 and geographic origin (Bouchet et al 2012 Morris et al 2013) which has the potential to 142 introduce false positive signals into GWAS analyses unless properly controlled (Yu et al 143 2006 Zhang et al 2010) In addition the average extent of linkage disequilibrium (LD) 144 decay in sorghum is substantially greater than in maize due at least in part to its mode of 145 reproduction (Chia et al 2012 Morris et al 2013) Therefore each trait-associated SNP 146 identified via GWAS is likely to be linked to a large genomic region making it challenging to 147 identify candidate genes However QTL studies have identified syntenic chromosomal 148 regions that control inflorescence traits in both maize and sorghum (Brown et al 2006) 149 Furthermore it is possible to identify conserved sorghum homologs of genes that regulate 150 inflorescence architecture in maize and rice (Paterson et al 2009 Schnable et al 2012 151 Zhang et al 2015) These findings suggest that scanning chromosomal regions surrounding 152 sorghum trait-associated SNPs for homologs of maize and rice genes with known functions 153 in inflorescence architecture could overcome the challenges in identifying candidate genes 154 caused by sorghumrsquos high LD155 In this study we developed and deployed a high-resolution imaging pipeline to collect 156 panicle phenotypes from a subset of the Sorghum Association Panel (SAP) (Casa et al 157 2008) consisting of 272 accessions We used a semi-automated procedure to extract both 158 uni- and multi-dimensional traits from these images The resulting phenotypic data were 159 used to perform GWAS which identified TASs some of which are located within 160 chromosomal regions identified via previously published QTL and GWAS analysis of 161 sorghum for similar traits In addition a statistically enriched fraction of these TASs are 162 located near sorghum homologs of maize or rice genes known to influence inflorescence 163 architecture Genome-wide analysis of population differentiation suggests that the genomic 164 regions that contain the TASs have undergone artificial selection during sorghum breeding 165

166


6

Results167

Phenotyping168 The front and side planes of 1064 panicles from 272 genotypes (designated SAP-FI 169 Supplemental Fig S1) were imaged (Fig 1 Materials and Methods) A MATLAB app from 170 TIM was used to mark the approximate boundary of each panicle that was subsequently 171 cropped from the original image A fully automated trait extraction protocol (Materials and 172 Methods) was then used to extract length and width directly from cropped front plane 173 images 174 To evaluate the accuracy of our semi-automated image processing method we manually 175 measured length and width of single panicles from 17 genotypes randomly selected from 176 the SAP-FI The resulting trait values were compared with the corresponding values 177 automatically extracted from images of the same panicles The coefficient of determination 178 (r2) between the values of the auto-extracted length and widthfront vs ground truth were 179 093 and 089 respectively (Fig 2) indicating that our pipeline can accurately extract 180 panicle length and width 181 Differences between the values for auto-extracted traits and ground truth as shown in Fig 2 182 consist of two components 1) variation between trait measurements of a 3D panicle and 183 the representation of these trait measurements in a 2D image and 2) errors in auto-184 extracting trait values from a 2D image To evaluate the first source of variation we 185 manually measured lengths from 2D panicle images of the above 17 genotypes using our 186 MATLAB app and compared the resulting trait values with panicle lengths of ground truth 187 The 096 r2 (Fig 3) indicated little variation was introduced via 2D imaging 188 To quantify the second source of variation we also manually measured lengths from 2D 189 images of panicles using our MATLAB app and compared the resulting trait values with 190 lengths auto-extracted from cropped images Considering all 1064 panicles the r2 was 094 191 (Fig 3) The r2 values could be further increased to 097 by comparing extracted trait values 192 of a genotype averaged across replications and locations (Fig 3) These results demonstrate 193 that the second source of error was minimal and it could be well controlled under our 194 experimental design Mean trait values were used for all subsequent phenotypic analyses 195 including GWAS 196 In addition to length and widthfront six other panicle traits were extracted from cropped 197 panicle images Widthside was extracted using the same criteria as widthfront but using a 198 paniclersquos side plane image Because the same pipeline was used to extract widthfront and 199 widthside we did not collect ground truth data for the latter trait Area and solidity are 200 features of a 3D object projected onto a 2D plane so we chose these two traits to reflect 201 structural variation in panicle size and compactness projected on the front and side planes 202


7

yielding the traits areafront areaside solidityfront and solidityside Panicle volume was 203 estimated using information extracted from both planes (Materials and Methods) No 204 ground truth data were collected for the five multi-dimensional traits due to the challenge 205 of measuring them by hand All eight traits exhibited high heritabilities (Fig 4) with the 206 highest for panicle length (h2=093) and lowest for widthside (h2=070) 207

PhenotypicandGeneticCorrelationsofPanicleTraits208 The phenotypic correlation between two traits is determined by their genetic and non-209 genetic correlations and heritabilities (Bernardo 2010) The genetic correlation between 210 two traits is a function of pleiotropy ie the effect of a gene on multiple traits (Mackay et 211 al 2009) Correlations among the eight traits were determined by analyzing all pair-wise 212 comparisons The phenotypic correlations among panicle traits were strongly associated 213


8

with their corresponding genotypic correlations (Fig 4) For example the ten pairs of traits 214 with the highest phenotypic correlations also exhibited the highest genetic correlations 215 For solidity and area values of the front plane were highly correlated with those of the side 216 plane (r values of 094 and 074 respectively Fig 4) In contrast as expected based on the 217 definition of front and side planes (Materials and Methods) there was only a low 218 correlation (r=018) between the panicle widths of each plane Although widthfront was 219 positively correlated with areafront (r=056) it was poorly correlated with areaside 220 (r=005) despite the reasonably high phenotypic correlation between the two area traits 221 (r=074) The genotypes within the SAP-FI panel exhibit a great deal of variation in panicle 222 architecture (Supplemental Fig S1) Some of these genotypes exhibit a flat shape in which 223 the widthside is much smaller than widthfront (see for example the three genotypes 224 displayed on the right of Supplemental Fig S1) Inspection of individual panicles confirmed 225 that flat genotypes contribute to the poor correlation between widthfront with areaside 226 (Supplemental Fig S2) Similarly both solidity traits were negatively correlated with traits 227 from the front plane (panicle length widthfront and areafront) while being positively 228 correlated with widthside and areaside Because these phenotypic correlations are 229 consistent with their genetic correlations (Fig 4) we hypothesize that some genes have 230 functions in both planes 231

GWAS232 To identify loci that affect panicle architecture GWAS was performed on the SAP-FI panel 233 for the eight panicle traits using an improved version of FarmCPU (Liu et al 2016) termed 234


9

FarmCPUpp (Kusmec and Schnable 2018) These analyses identified 49 associations 235 (Supplemental Fig S 3) representing 46 non-redundant trait-associated SNPs (TASs) A 236 bootstrapping process was used to filter false positive signals (Materials and Methods) Of 237 the 46 SNPs detected in the GWAS of the SAP-FI panel 38 associations representing 35 238 high confidence non-redundant TASs also exhibited a resample model inclusion probability 239 (RMIP) ge 005 in the bootstrap experiment (Methods Table I) 240 Sixty-six percent (N=23) of these 35 high-confidence non-redundant TASs were associated 241 with three of the eight traits length (N=6) solidityfront (N=8) and widthfront (N=9) For 242 the three traits measured in both planes approximately twice as many TASs were detected 243 for front plane traits (N=21) as compared to the corresponding side plane (N=10) (Table I) 244 This probably reflects the lower heritabilities of the side plane traits relative to the front 245 plane (Fig 4) 246 We compared the TASs associated with panicle length and widthfront with published QTL 247 and GWAS results (Hart et al 2001 Brown et al 2006 Srinivas et al 2009 Morris et al 248 2013 Nagaraja Reddy et al 2013 Zhang et al 2015 Zhao et al 2016) Encouragingly two 249 of our six TASs for panicle length are located within previously reported QTL intervals 250 (Supplemental Table S1 Hart etal 2001 Srinivas etal 2009) and another (at position 251 8065027 bp on chromosome 1) is located within the interval reported by a previous GWAS 252 for panicle length (Zhang et al 2015) One of the nine TASs identified for widthfront (at 253


10

position 3724913 bp on chromosome 3) was located within regions identified via previous 254 QTL and GWAS on panicle width and branch length (Hart et al 2001 Brown et al 2006 255 Morris et al 2013 Zhang et al 2015) Given the differences in genetic materials used in 256 these studies and the fact that SAP-FI is only a subset of the original SAP used for GWAS by 257 others it is not surprising that the results of these studies do not fully overlap 258


11

Only three TASs were associated with multiple traits two were associated with both 259 solidityfront and solidityside (Table I) and one with both areaside and volume The limited 260 number of overlapping SNPs was unexpected given the substantial correlations among 261 traits (Fig 4) We hypothesized that this low overlap is due to false negative associations 262 that arose as a consequence of the modest effect sizes of many of the detected TASs 263 (standardized effect sizes lt 05 Supplemental Fig S4) in combination with the relatively 264 small size of our panel which limited the power of statistical tests To evaluate this 265 hypothesis we compared the high-confidence TASs for one trait with all significant TASs 266 identified via bootstrapping that exhibited an RMIP ge 005 for a second trait highly 267 correlated with the first one This process identified nine additional TASs (Supplemental 268 Table S2) affecting highly correlated traits supporting our hypothesis that our stringent 269 control of false-positive signals reduced the ability to detect TASs that associated with 270 multiple traits Similar to previous studies on the effect sizes of pleiotropic TASs for maize 271 inflorescences (Brown et al 2011) the standardized effect sizes of the twelve sorghum 272 TASs that affect multiple traits were similar for both members of pairs of affected traits 273 (Supplemental Fig S5) 274

IdentificationofCandidateGenes275 The sorghum genome exhibits extensive LD (Morris et al 2013) Although estimates of LD 276 vary across the genome (Hamblin et al 2005 Bouchet et al 2012 Morris et al 2013) we 277 elected to use a genome-wide average estimate of LD in our search for candidate genes 278 because our local estimates of LD surrounding TASs were noisy perhaps due to the 279 relatively small number of SNPs per Mb (200=146865730Mb genome) and missing data 280 in this GBS-derived SNP set Hence we screened 700-kb windows centered on each TAS for 281 candidate genes (Material and Methods) 282 The limited number of sorghum genes that have been subjected to functional analyses 283 complicates the identification of candidate genes We overcame this challenge by looking 284 for sorghum homologs of maize and rice genes (Supplemental Table S3) previously 285 associated with inflorescence architecture (Vollbrecht et al 2005 Tanaka et al 2013 286 Zhang and Yuan 2014) located within the 700-kb windows surrounding TASs (Materials 287 and Methods) 288 Using this procedure nine candidate genes were identified (Table II) A permutation test 289 (Materials and Methods p-value = 0026) indicated that sorghum homologs of maize and 290 rice genes known to affect inflorescence architecture are enriched in chromosomal regions 291 surrounding TASs In addition the maize and rice homologs of each of the nine sorghum 292 candidate genes have functions consistent with the traits used to identify associations 293 (Table II) and five of these maize andor rice genes were associated with relevant traits 294 based on GWAS conducted in those species (Table II) 295


12

PopulationDifferentiation296 Because inflorescence traits can influence per plant yield these traits have likely been 297 targets of selection According to Casa et al (2008) the SAP-FI (N=272) used in this study 298 contains 55 elite inbred lines and landraces that were developed in the United States 40 299 cultivars collected worldwide and 177 converted tropical lines The latter group was 300 generated by the Sorghum Conversion Program (SCP) which aimed to introduce novel 301 genetic variation from exotic tropical germplasm collected from around the world into 302 modern US breeding lines (Stephens et al 1967) During this process tropical lines were 303 backcrossed to a single adapted line (BTx406) and selection was performed only for 304 flowering time and plant height Specifically there was no direct selection for panicle 305 architecture and thus much of the natural genetic variation present in the tropical lines 306 associated with panicle traits would have presumably been retained in the converted 307 tropical lines (Casa et al 2008 Thurber et al 2013) As a group the 177 converted tropical 308 lines have shorter and narrower panicles as well as smaller cross-sectional areas compared 309 to the 40 cultivars (Supplemental Fig S6) The origins of the 40 cultivars and exotic donors 310 of the 177 converted tropical lines have similar geographic distributions (Casa et al 2008) 311 Hence by identifying chromosomal regions that exhibit statistically significant differences 312 in allele frequencies between the two groups (Materials and Methods) we can identify 313 regions that have putatively been under selection We could then ask whether such 314 chromosomal regions are enriched for our TASs 315 A prior comparison of the parental and converted tropical lines generated in the SCP 316 identified chromosomal regions on the long arms of chromosomes 6 7 and 9 that exhibit 317 signatures of selection Consistent with the nature of the SCP these regions contain clusters 318 of genes (Ma1 Ma6 Dw1 Dw2 and Dw3) that regulate plant height and flowering time 319 (Thurber et al 2013) Our comparison between the 177 converted tropical lines and 40 320 cultivars also identified the regions (Fig 5) associated with flowering time and plant height 321 described by Thurber et al (2013) At first this result was surprising but subsequent 322 analyses on previously reported plant height loci (Brown et al 2008 Li et al 2015) 323 revealed that the 40 cultivars have lower frequencies of the favorable alleles of three height 324 genes located on chromosomes 6 7 and 9 that were contributed by BTx406 than the 325 converted tropical lines 326


13

In addition to the adaptation genes located on chromosomes 6 7 and 9 we also identified 327 genomic regions that exhibit signatures of selection that were not identified by Thurber et 328 al (2013) (Fig 5) These regions probably reflect selection during modern breeding for 329 agronomic traits potentially including panicle architecture Consistent with the hypothesis 330 that selection has occurred for panicle traits 26 of our 35 TASs co-localized within 331 chromosomal regions that exhibit evidence of selection Eight of these 26 TASs are linked to 332 candidate genes (Fig 5) Based on permutation tests more overlap was observed than 333 would be expected if the TASs had not been under selection (P value lt 0001) 334 To further test this hypothesis we determined trait values for areafront for each of the four 335 potential genotypes associated with the two TASs for this trait both of which exhibited 336 signatures of selection For the first TAS (S1_3176715) the reference (R) allele is associated 337 with larger areafront In contrast the alternative (A) allele of the second TAS 338 (S4_12278584) is associated with larger areafront Considering all members of the SAP-FI 339 the RA and AR genotypes have the largest and smallest areafront values respectively (Fig 340 6) Consistent with our hypothesis that differences in allele frequencies of these two TASs in 341 the two groups is due to selection for panicle architecture the favorable R allele of SNP 342 S1_3176715 exhibits a higher (29) frequency in the cultivars (2640) than the converted 343 tropical lines (89177) The frequency of the favorable A allele of SNP S4_12278584 is not 344


14

higher in the cultivar than the converted tropical lines only one of the 40 cultivars carries 345 the least favorable genotype of both SNPs (AR) Although not statistically significant 346 (Fisherrsquos Exact Test P value = 0069) this frequency (140) is five times lower than that 347 observed in the converted tropical lines (20177) (Fig 6) Similarly the frequency of the 348


15

least favorable genotypic class determined by the two SNPs associated with panicle length 349 was reduced in cultivars as compared to the converted tropical lines (Fig 6) These results 350 at least suggest the possibility that chromosomal regions surrounding some of the TASs 351 identified in this study may have been targets of selection during modern breeding 352

353


16

Discussion354 Given the substantial reductions in the cost of genotyping large panels phenotyping has 355 become the bottleneck for large-scale genetic analyses of crops Two groups have reported 356 pipelines (P-TRAP and PANorama) to capture phenotypes of rice panicles using 357 photographs of flattened panicles (AL-Tam et al 2013 Crowell et al 2014) In both cases 358 only a single image was analyzed per flattened panicle Hence these pipelines are not 359 suitable for extracting multi-dimensional traits associated with the 3D structures of 360 panicles Others have used complex imaging chambers to measure 3D structures of limited 361 numbers of Arabidopsis inflorescences (Hall and Ellis 2012) but it would be difficult to 362 scale this approach to phenotype large diversity panels 363 Here we present a low-cost easy to replicate pipeline that was used to phenotype panicles 364 in a sorghum diversity panel To explore variation in panicle architecture among the 365 members of this diversity panel we captured images from the two planes of a panicle This 366 enabled us to extract not only unidimensional traits such as panicle length and width but 367 also multi-dimensional traits such as areafront areaside solidityfront solidityside and 368 volume Extracted trait values for panicle length and width exhibited high accuracies (093 369 and 089 respectively) as compared to ground truth measurements of intact panicles 370 Because it is not possible to extract ground-truth data for multi-dimensional traits we 371 could not determine the accuracy of our pipeline for these traits Even so since the 372 heritabilities were high (070-093) we conducted a GWAS for these traits and identified 373 candidate genes associated with some of the TASs 374 Given the extensive LD present in sorghum it is not possible to conclude that any particular 375 candidate gene affects the associated trait However the enrichment of sorghum homologs 376 for maize and rice genes known to affect inflorescence architecture near these TASs and 377 the correspondence of their functions in maize and rice with the associated phenotypes in 378 sorghum (Table II) support the accuracy of our phenotyping pipeline and the hypothesis 379 that at least some of the candidate genes are causative 380 Functional studies (Tanaka et al 2013 Zhang and Yuan 2014) have demonstrated that 381 many genes that regulate inflorescence development are functionally conserved among 382 grass species More specifically GWAS suggest the conservation of genetic architecture 383 between rice panicles and maize tassels (Brown et al 2011 Crowell et al 2016 Xu et al 384 2017) Hence we prioritized our selection of candidates by identifying sorghum homologs 385 of inflorescence genes previously discovered in maize and rice In this manner we 386 identified nine candidate genes most of which are related to Kn1 (Fig 7) 387 In maize both Kn1- and Ra1-related genes affect tassel development and many auxin-388 related genes are targeted by both the RA1 and KN1 proteins (Vollbrecht et al 2005 389 McSteen 2006 Eveland et al 2014) Although our GWAS associated many Kn1-related 390


17

sorghum genes with panicle architecture no TASs were found in LD with members of the 391 ramosa pathway This was surprising because in maize this pathway contributes to the 392 formation of spikelet pair meristems and has been associated with tassel architecture via 393 GWAS (Brown et al 2011 Wu et al 2016 Xu et al 2017) Given the level of functional 394 conservation between these two species it is unlikely that the ramosa pathway does not 395 contribute to the development of sorghum panicles Instead the failure of our GWAS to 396


18

detect associations with genes in the ramosa pathway suggests that differentially functional 397 alleles of these genes are not segregating at high frequencies in the SAP-FI This finding 398 highlights the value of combining high-throughput phenotyping with GWAS to identify 399 standing variation that can be used to select targets for marker-assisted breeding 400 On average cultivars have larger panicles than converted tropical lines and our results 401 suggest that by employing phenotypic selection breeders have selected for favorable 402 alleles of genes that regulate panicle architecture Even so our results also indicate that 403 cultivars have not been purged of unfavorable alleles of these genes This is likely because 404 most TASs identified in this study exhibit only moderate effect sizes and consequently and 405 consistent with expectation (Bernardo 2008 L Heffner et al 2009 Bernardo 2016) 406 traditional phenotypic selection has not yet fixed favorable alleles at all relevant loci This 407 presents an opportunity to employ the TASs identified in our GWAS to design a marker-408 assisted selection program to improve panicle architecture traits with impact on final grain 409 yield 410 The general imaging and data analysis approach reported here could be used to phenotype 411 other grass species with panicle structures similar to those of sorghum Often academic 412 software is not supported on the latest versions of operating systems Our code is based on 413 built-in MATLAB functions which does not put the burden of long-term maintenance on the 414 developers or users Hence anyone in possession of our scripts can run them on MATLAB 415 which is commercially maintained and routinely updated to cater to all operating systems 416 In addition our semi-automated trait extraction pipeline requires no advanced hardware 417 any commercially available laptop is capable of completing trait extractions from 1000 418 images within 24 hours As such it should be possible to deploy this method even at remote 419 breeding stations in developing countries 420

MaterialsandMethods421

ImagingPanicles422 Three hundred and two sorghum genotypes from the sorghum association panel (SAP) 423 (Casa et al 2008) were grown at two Iowa State University experimental farms (Kelley 424 Farm in Ames IA and Burkey Farm in Boone IA) in 2015 This panel was planted in a 425 randomized complete block design with two replications at each location as previously 426 described by Salas Fernandez etal (2017) One panicle per replicate per location was 427 harvested during the third week of October and imaged using a light box A few genotypes 428 had some panicles with drooping primary branches To avoid artifacts due to these 429 drooping panicles we selected for phenotyping within such genotypes only those specific 430 panicles that did not droop The light box was constructed with a ldquoPacific Bluerdquo background 431 to ensure easy segmentation of panicles in subsequent image processing steps This box 432 was designed to accommodate the diversity of panicles in the SAP (Figure 1) with 433


19

dimensions of about 45 cm tall 36 cm wide and 24 cm deep All images were captured 434 using a single Canon 5DS DSLR camera with EF100 mm f28L Macro IS USM lens The 435 camera was mounted at a fixed height and angle to ensure consistent imaging 436 Each panicle was mounted upright in the center of the light box with a reference scale of 437 known dimensions placed next to it The reference scale allows automated conversion of 438 size dependent traits from the number of pixels occupied in the image to real lengths ie 439 inches or centimeters The reference scale was initially a 65 cm wide white scale but was 440 later changed to a 254 cm (1 inch) wide yellow scale for easier extraction from the blue 441 background Because the images were 2D projections of the panicles the panicles were 442 imaged along two perpendicular planes based on a front view and a side view The plane 443 that exhibited the largest cross-sectional area ndash as determined via visual inspection ndash was 444 designated the front plane and the plane perpendicular to it as the side plane The front 445 plane of a panicle was always imaged first 446

ExtractionofTraitsfromImages447 We developed a pipeline TIM (Toolkit for Inflorescence Measurement available at the 448 Schnable Labrsquos GitHub page httpsgithubcomschnablelab) to semi-automatically 449 extract traits from panicle images The trait extraction via TIM involved the following steps 450 i) segmentation of the panicle and the reference scale from the background using a custom-451 built MATLAB app (Supplementary Methods) ii) measurement of traits on the segmented 452 panicle and iii) conversion of trait values from pixels to metric measurements using a 453 reference scale (Fig 1 Supplementary Methods) Trait measurement was performed in a 454 fully automated manner using standard image processing algorithms (described in 455 Supplementary Methods) Image processing is particularly suited to traits that can be 456 cumbersome to measure by hand such as the exact cross-sectional area of the panicle 457 solidity etc However some traits such as the length and width can be challenging to 458 extract via automated image processing from strongly curved panicles whose tips touched 459 the bottom of the light box making it difficult to distinguish the first internode of a 460 paniclersquos rachis Therefore we removed 30 genotypes that consistently exhibited u-shaped 461 panicles across locations and reps thus 272 genotypes and 1064 panicles were 462 subsequently used for image processing and data analyses Due to missing data the total 463 number of imaged panicles was 1064 rather than 1188 (=272 x 2 repslocation x 2 464 locations) In this report this subset of the SAP (N=272 genotypes) will be referred to as 465 SAP-FI 466

TraitExtractionEight panicle traits were extracted from front and side plane images of the 467 1064 panicles as described below These are length width area volume and solidity Of 468 these width area and solidity were extracted from both the front and side planes Length is 469 defined as the length of the main panicle axis from the region of the lowest branch point to 470


20

the tip of the panicle Because length is expected to be the same in the two planes it was 471 extracted from only the front plane Volume is a derived quantity that is estimated jointly 472 from the front and side plane images The front plane and side plane images were first 473 matched to each other height-wise Subsequently the widths from the front plane and side 474 plane were extracted along the length of the panicle in slices from the top to the bottom (S 475 Methods) Assuming the maximum and minimum widths represent the major axis and the 476 minor axis of an ellipse it is possible to calculate the elliptical cross-sectional area (an 477 ellipse being an approximation of the shape of the panicle) Integrating the cross-sectional 478 area over the height of the panicle provides an approximate volume of the panicle 479 The remaining traits width area and solidity were extracted from both the front and side 480 plane images (widthfront widthside areafront areaside solidityfront and solidityside) 481 Width represents the maximum value of the width of the panicle obtained from the longest 482 line that can be drawn between the two boundaries of the panicle and that is perpendicular 483 to the paniclersquos length (Fig 1) Area refers to the projected area of the panicle on a 2D plane 484 because the panicle is a 3D structure Separate area measurements were obtained from the 485 number of pixels contained within the projections of the panicle on the front and side 486 planes Solidity is a measure of whether the panicle is loosely or tightly packed It is derived 487 from the convex hull which is the smallest polygonal bounding curve that encompasses the 488 shape of the panicle Solidity is the ratio of the Projected Area of the panicle to the area of 489 its convex hull 490 With the exception of solidity ndash which is a dimensionless quantity ndash all trait values were 491 measured in units of pixels The pixel-to-cm conversion was obtained from the pixel width 492 of the 1rdquo yellow scale in each image (pixelcm = pixel width of yellow scale254) to obtain 493 the actual lengths of these panicles Phenotype values of every trait were then transformed 494 into standard units (cm for length and width cm2 for area cm3 for volume) based on the 495 pixelcm ratio calculated for each image 496

AnalysisofPhenotypicData497 For each phenotype variance components were estimated using the lme4 package (Bates 498 et al 2014) in R (version 342) using the following function lmer(phenotype~(1|genotype) 499 +(1|location) + (1|LocationRep) + (1| genotype location)) Entry mean-based heritabilities 500 (H2) were calculated from variance components estimates (Bernardo 2010) as follows H2 501 = VG[VG + (VGEn) + (Ve(nr)] where VG is the genotypic variance VGE is the variance of 502 genotype by location interaction Ve is the residual variance n is the number of locations 503 and r is the number of replications Phenotypic correlations among eight traits were 504 estimated using Pearson correlations between two traits using the mean value of each 505 genotype via the lsquocorrsquo function of R 506

GenotypingData507


21

We used SNP data generated and imputed by Morris et al 2013 These SNPs were filtered 508 to retain only those with a minor allele frequency (MAF) gt005 and a missing rate lt60 in 509 the SAP-FI The remaining 146865 SNPs were used to calculate genetic correlations among 510 the eight panicle traits using bivariate genomic-relatedness-based restricted maximum-511 likelihood (GREML) analysis from GCTA (Yang et al 2010 Yang et al 2011) 512

GWAS513 GWAS was conducted using the 146865 SNPs in R using a version of FarmCPU (Liu et al 514 2016) termed FarmCPUpp which was modified to increase computational efficiency and 515 reduce run times (Kusmec et al 2017) FarmCPU combines SNP-based covariate selection 516 (MLMM Segura et al 2012) and restricted kinship matrices (SUPER Wang et al 2014) to 517 reduce false positives and negatives Simulations demonstrate that FarmCPU achieves both 518 of these goals in addition to increasing computational efficiency (Liu et al 2016) PCA was 519 conducted on the SNP data using the R lsquoprcomprsquo function The first three principal 520 components were used as covariates to control for population structure FarmCPUpprsquos 521 optimum bin selection procedure was conducted using bin sizes of 50 kb 100 kb and 500 522 kb and pseudo-quantitative trait nucleotide values of 3 6 9 12 and 15 Statistical 523 significance was determined after Bonferroni correction α= (005total number of SNPs) 524 Initially GWAS was conducted using all 272 accessions of the SAP-FI (Supplemental Table 525 S4) To identify high-confidence SNPs bootstraps were conducted for each phenotype 526 following previously described methods (Brown et al 2011 Wallace et al 2014) The full 527 panel was first divided into four sub-panels determined via fastSTRUCTURE (Raj et al 528 2014) Then we performed 100 bootstraps with each iteration randomly assigned 10 of 529 the phenotypic values within each sub-panel as missing Then for each iteration GWAS was 530 repeated using the same parameters as for the full panel The RMIP was calculated based on 531 the fraction of bootstraps in which a SNP was significantly associated with the phenotype 532 Only SNPs with RMIP ge5 (5 out of 100 iterations) were considered for further analysis We 533 also made this process more stringent by requiring these hits to be significant in the GWAS 534 conducted on the full panel These SNPs were defined as high-confidence trait-associated 535 SNPs (TASs) and used to screen for candidate genes 536

Co-localizationofTrait-associatedSNPsandCandidateGenes537 Genome-wide LD was assessed using PLINK (version 19 wwwcog-538 genomicsorgplink19 Chang etal 2015) by calculating the pairwise r2 of every SNP 539 within 1 Mb using 25 kb steps The average length of the interval within which r2 fell below 540 01 was 350 kb Hence we scanned 350 kb upstream and downstream of each TASs for 541 candidate genes 542


22

Potential candidate genes were first identified by screening the literature for cloned maize 543 and rice genes with known functions in inflorescence architecture (Supplemental Table S3) 544 The corresponding sorghum homologs based on sorghum genome V14 from PlantGDB 545 (httpwwwplantgdborgSbGDB) of the resulting 67 maize and 17 rice genes were 546 identified using MaizeGDB (Schnable et al 2012) and the rice genome annotation project 547 (Kawahara et al 2013) respectively To test whether the distribution of candidate genes 548 was enriched in regions surrounding the TASs we performed 1000 permutations during 549 which 35 SNPs (equal to the number of TASs detected via GWAS) were randomly selected 550 from the 146865 SNPs for each permutation We then recorded the number of 551 permutations in which nine or more candidate genes were present within 700 kb windows 552 surrounding the randomly sampled SNPs This process was repeated 10 iterations and the 553 median p-value from these iterations was used to evaluate the significance554

PopulationDifferentiation555 One hundred and seventy-seven converted tropical lines and 40 cultivars from the SAP-FI 556 were used to analyze genome-wide population differentiation Genotypic data of the two 557 sub-panels were first filtered to have a missing rate of lt5 within each sub-panel The 558 Hudson estimator of Fst (Hudson et al 1992) was calculated using the 74142 filtered SNPs 559 shared by both sub-panels to estimate the genome wide divergence between the two 560 following previously described methods (Bhatia et al 2013) 561

AccessionNumbers562 Sequence data used in this paper can be found in the Sequence Read Archive database 563 wwwncbinlmnihgovsra (accession no SRA062716) 564

SupplementalData565 Supplemental Figure S1 Example showing the structural diversity of SAP-FI 566 Supplemental Figure S2 Phenotypic correlation between widthfront and areaside with 567 examples showing genotypesrsquo panicle shapes that are circled in the scatter plot 568 Supplemental Figure S3 Manhattan and QQ plots for eight panicle traits 569 Supplemental Figure S4 Standardized effect sizes of 38 TASs 570 Supplemental Figure S5 Standardized effect sizes of SNPs associated with multiple traits 571 Supplemental Figure S6 Boxplot of values of eight panicle traits in 177 converted tropical 572 lines and 40 cultivars 573


23

SupplementalTableS1 TASs that overlapped with previously reported QTL and GWAS 574 intervals on panicle length panicle width and branch length 575

SupplementalTableS2 List of TASs with pleotropic effect in 100 bootstrapping 576

SupplementalTable S3 List of sorghum homologs of maize or rice genes with known 577 function on inflorescence architecture 578

Supplemental Table S4 List of mean phenotypic value of eight panicle traits of 272 579 genotypes used in GWAS study 580 581

Acknowledgements582 We thank Ms Lisa Coffey (Schnable laboratory) and Ms Nicole Lindsey (Salas Fernandez 583 laboratory) for assistance with designing and conducting the sorghum field experiments 584 graduate student Zihao Zheng (Schnable laboratory) for assistance harvesting panicles and 585 undergraduate students Stephanie Shuler Jodie Johnson and Katherine Lenson (Schnable 586 laboratory) for assistance collecting image data 587 588 589 590 591 592 593 594 595 596 Table I List of 35 significant TASs affecting 38 trait associations Trait SNP -log10(P value) Estimate effect Standardized effect size RMIP

Areafront S1_3176715 980 -575 027 047 Areafront S4_12278584 b 656 -416 020 013 Areafront S4_60591319 688 -538 025 019 wwwplantphysiolorgon March 3 2020 - Published by Downloaded from

Copyright copy 2018 American Society of Plant Biologists All rights reserved

24

Areafront S10_1793055 658 496 023 022 Areaside S3_3201555c 947 655 051 016 Areaside S3_58994806 b 807 353 028 012 Areaside S5_17518239 674 407 032 044 Areaside S6_49755954 791 -328 026 030 Panicle Length S1_8065027 a b 669 098 022 007 Panicle Length S2_63553713 a b 1521 336 074 075 Panicle Length S3_58920501 b 692 090 020 005 Panicle Length S8_14556347 a 728 -167 037 040 Panicle Length S10_50809289 1068 131 029 019 Panicle Length S10_54184063 688 -084 018 017 Solidityfront S2_10933862C 1507 -007 071 082 Solidityfront S2_70599334 1056 -003 032 009 Solidityfront S2_76284762 b 948 004 041 005 Solidityfront S3_56132328 659 002 016 027 Solidityfront S4_56171463 1534 -003 036 045 Solidityfront S6_60219786 b 739 002 026 007 Solidityfront S8_6056282c 740 -003 036 013 Solidityfront S10_1149373 662 -002 024 005 Solidityside S1_19664643 b 720 -004 059 031 Solidityside S2_10933862c 1013 -005 066 081 Solidityside S7_54381792 913 004 056 031 Solidityside S8_6056282c 757 -003 041 038 Volume S3_3201555c 934 3333 053 045 Widthfront S2_68297155 1417 053 038 072 Widthfront S3_3724913 887 048 034 048 Widthfront S3_57037006 1099 071 050 069 Widthfront S4_65197133 761 053 038 010 Widthfront S5_11264509 798 058 041 015 Widthfront S6_60559938 662 042 030 017 Widthfront S8_26248649 804 058 042 018 Widthfront S8_45726712 935 044 031 012 Widthfront S9_3027904 781 062 044 024 Widthside S3_2724080 b 957 027 043 019 Widthside S10_60236042 b 682 015 023 008 aTASs that overlap with previously reported QTLs or GWAS intervals (Supplemental Table S1) bHave candidate gene located within 350 kb on either side of the TAS (Table II) cPleiotropic TASs in GWAS of full set


25

Table II List of nine candidate genes for panicle architecture Traits SNP ID Sorghum Candidate Gene ID (V14)

Annotation of Sorghum Candidate Gene Distance from SNP to Candidate Gene (Kb)

Homolog of Candidate gene in Maize (Zm) or Rice (Os) Function of Homolog in MaizeRice Inflorescence Development

Traits Associated with Homolog Gene via GWAS in MaizeRice References

Panicle Length S1_8065027 Sb01g009480 Homeobox domain containing protein 1576 Kn1(Zm) Meristem maintenance and regulate transition from SAM into inflorescence meristem

Central Spike Length Vollbrechtetal2000Brownetal2011Xuetal2017

Solidityside S1_19664643 Sb01g018890 Dicer-like 3 1443 dicer-like3(DCL3Os)

Regulating GA and BR homeostasis Mutation cause reduced secondary branches

Weietal2014

Panicle Length S2_63553713 Sb02g028420 SBP-box gene family member 430 tsh4(Zm) Lateral meristem initiation and stem elongation Tassel Length Chucketal2010

Brownetal2011Xuetal2017Solidityfront S2_76284762 Sb02g042400 AP2 domain containing protein 1461 BD1(Zm) Specification of the spikelet meristem identity and inhibit the formation of axillary meristems

Secondary Branching Branch Length Chucketal2002

Brownetal2011

Crowelletal2016Wuetal2016Xuetal2017Widthside S3_2724080 Sb03g002525 MADS-box family gene with MIKCc type-box

3078 MADS3(Os)zmm2(Zm)

Specification of the spikelet meristem identity Yamaguchietal2006

w

ww

plantphysiolorgon M

arch 3 2020 - Published by

Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26

Areaside S3_58994806 Sb03g030635 Homeobox-3 1428 DWT1(Os) Stem cell elongation and cell proliferation Wangetal2014Panicle Length S3_58920501 Sb03g030635 Homeobox-3 685 DWT1(Os) Stem cell elongation and cell proliferation Wangetal2014Areafront S4_12278584 Sb04g009700 Galactose oxidasekelch repeat superfamily protein

33 Largerpanicle(LP

Os) Modulating cytokinin level in inflorescence and branch meristem Panicle Size Panicle Length Lietal

2011Crowelletal2016Wuetal2017Solidityfront S6_60219786 Sb06g031880 Homeobox domain containing protein

402 WUS(Os) Inhibiting differentiation of the stem cells NardmannampWerr2006Widthside S10_6023604

2Sb10g030270 Receptor protein kinase CLAVATA1 precursor

2731 Td1(Zm) Regulating shoot and floral meristem size Cob Diameter Kernel Row Number Bommertetal2005

Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598

Figure1Phenotyping pipeline used to extract traits from 1064 panicles of 272 genotypes 599 Each panicle was photographed from both the front and side planes Segmentation was 600 conducted manually for subsequent automatic trait extraction 601

Figure 2 Comparison of ground-truth measurements and trait values extracted from 602 images of 17 randomly chosen panicles A Panicle length B Panicle width of front plane 603

Figure3Comparison of panicle lengths manually measured from images using a custom 604 MATLAB app and extracted from the same images using our automated image processing 605 pipeline A Comparison of ground-truth measurements and panicle length manually 606 measured from images of 17 randomly chosen panicles (front plane) B Panicle lengths 607 from 1064 individual images (front plane) C Panicle lengths using mean values of 272 608 genotypes 609

Figure 4 Phenotypic and genetic correlations among panicle traits Traits are ordered 610 using the lsquohclustrsquo function of R with phenotypic and genetic correlations indicated by 611 ellipses and squares respectively Traits extracted from front or side plane are 612 distinguished by the extensions ldquofrontrdquo and ldquosiderdquo respectively The heritability of each trait 613 is shown on the left 614

Figure5 Identification of chromosomal regions that exhibit evidence of selection Allele 615 frequencies were compared between 177 converted tropical lines and 40 cultivars A 995 616 Fst cut-off is indicated by horizontal dashed lines Vertical dashed lines designate positions 617 of 35 TASs TASs located within divergent regions are indicated in red Two TASs on 618 chromosome 3 at 58920501 bp and 58994806 bp are represented by a single red dashed 619 line Candidate genes are labeled Gray shading on chromosomes 6 7 and 9 indicates 620 genomic regions identified by Thurber et al (2013) carrying maturity and plant height 621 genes introduced in the SCP 622

Figure6 Evidence of selection against alleles for small panicle areas and short panicles 623 Allele from the reference genome of BTx623 is designated as reference (R) allele and other 624 alleles are designated as alternative (A) allele A and B) panicle area C and D) panicle 625 length Phenotypes associated with all four genotypes of TASs S1_3176715 and 626 S4_12278584 (A) and TASs S1_8065027 and S3_58920501 (C) in the complete panel 627 Genotype frequencies of TASs in the converted tropical lines and cultivars (B and D) 628

Figure 7 Potential relationships among nine candidate genes (plus ga20ox) and their 629 functions in inflorescence development Kn1 is a key regulator of floral transition 630 (Vollbrecht etal 2000 Zhang amp Yuan 2014) that is essential for meristem maintenance 631 and is involved in the initiation of spikelet and floral meristems in rice and maize 632 (Kerstetter etal 1997 Tsuda etal 2011) Analyses of maize mutants demonstrate that 633 Kn1 regulates the cytokinin (CK) pathway perhaps via interactions with Td1 a suppressor 634


28

of the WUS gene (Bommert etal 2005 Lunde amp Hake 2009) Kn1 also participates in 635 regulating gibberellin (GA) levels in the shoot apical meristem through ga20ox and various 636 KNOTTED1-like homeobox (KNOX) proteins (Bolduc amp Hake 2009) In sorghum the 637 expression of Kn1 is strongly correlated with DCL-3 which is involved in the production of 638 24-nt siRNAs that target genes (eg ga20ox) involved in homeostasis of the GA and 639 brassinosteroid (BR) pathways to alter branch meristems in rice (Wei etal 2014) In 640 addition GWAS on panicle area identified one WUSCHEL-related homeobox transcription 641 factor DWT1 which in rice is associated with GA signaling participates in panicle meristem 642 elongation and is potentially downstream of ga20ox (Wang etal 2014) In maize tsh4 643 regulates the establishment of lateral meristem boundaries followed by meristem initiation 644 and is regulated via the expression of Kn1 and Bd1 (Chuck etal 2010) In rice and maize 645 Kn1 participates in spikelet meristem initiation and floral meristem initiation which are 646 positively regulated by Bd1 and MADS3 (Kerstetter et al 1997 Colombo et al 1998 647 Yamaguchi etal 2006 Bolduc amp Hake 2009)648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665


29

666 667 668 669


Parsed CitationsAL-Tam F Adam H Anjos A Lorieux M Larmande P Ghesquiegravere A Jouannic S Shahbazkia H (2013) P-TRAP a Panicle TraitPhenotyping tool BMC Plant Biol 13 122

Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Amine EK Baba NH Belhadj M Deurenberg-Yap M Djazayery A Forrestre T Galuska DA Herman S James WPT MBuyambaKabangu JR (2003) Diet nutrition and the prevention of chronic diseases World Health Organ Tech Rep Ser


Aquino A Millan B Gaston D Diago MP Tardaguila J (2015) vitisFlowerreg Development and testing of a novel android-smartphoneapplication for assessing the number of grapevine flowers per inflorescence using artificial vision techniques Sensors 15 21204ndash21218


Bates D Maumlchler M Bolker B Walker S (2014) Fitting Linear Mixed-Effects Models using lme4 doi 1018637jssv067i01Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Bernardo R (2010) Breeding for Quantitative Traits in Plants 2nd ed 2nd ed Stemma Press Woodbury MNPubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Bernardo R (2008) Molecular markers and selection for complex traits in plants Learning from the last 20 years Crop Sci 48 1649ndash1664Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Bernardo R (2016) Bandwagons I too have known Theor Appl Genet 129 2323ndash2332Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Bhatia G Patterson N Sankararaman S Price a L (2013) Estimating and interpreting FST The impact of rare variants Genome Res 231514ndash1521


Bolduc N Hake S (2009) The Maize Transcription Factor KNOTTED1 Directly Regulates the Gibberellin Catabolism Gene ga2ox1Plant Cell 21 1647ndash1658


Bolduc N Yilmaz A Mejia-Guerra MK Morohashi K OConnor D Grotewold E Hake S (2012) Unraveling the KNOTTED1 regulatorynetwork in maize meristems Genes Dev 26 1685ndash1690


Bommert P Lunde C Nardmann J Vollbrecht E Running M Jackson D Hake S Werr W (2005) thick tassel dwarf1 encodes a putativemaize ortholog of the Arabidopsis CLAVATA1 leucine-rich repeat receptor-like kinase Development 132 1235ndash1245


Bouchet S Pot D Deu M Rami JF Billot C Perrier X Rivallan R Gardes L Xia L Wenzl P et al (2012) Genetic structure linkagedisequilibrium and signature of selection in sorghum Lessons from physically anchored DArT markers PLoS One doi101371journalpone0033470


Brown PJ Klein PE Bortiri E Acharya CB Rooney WL Kresovich S (2006) Inheritance of inflorescence architecture in sorghumTheor Appl Genet 113 931ndash942


Brown PJ Rooney WL Franks C Kresovich S (2008) Efficient mapping of plant height quantitative trait loci in a sorghum associationpopulation with introgressed dwarfing genes Genetics 180 629ndash637


Brown PJ Upadyayula N Mahone GS Tian F Bradbury PJ Myles S Holland JB Flint-Garcia S McMullen MD Buckler ES et al (2011)Distinct genetic architectures for male and female inflorescence traits of maize PLoS Genet doi 101371journalpgen1002383

Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title wwwplantphysiolorgon March 3 2020 - Published by Downloaded from


Bruinsma J (2017) World agriculture towards 20152030 an FAO study RoutledgePubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Buckler ES Holland JB Bradbury PJ Acharya CB Brown PJ Browne C Ersoz E Flint-Garcia S Garcia A Glaubitz JC et al (2009) TheGenetic Architecture of Maize Flowering Time Science 325 714ndash718


Casa AM Pressoir G Brown PJ Mitchell SE Rooney WL Tuinstra MR Franks CD Kresovich S (2008) Community resources andstrategies for association mapping in Sorghum Crop Sci 48 30ndash40


Chang CC Chow CC Tellier LCAM Vattikuti S Purcell SM Lee JJ (2015) Second-generation PLINK Rising to the challenge of largerand richer datasets Gigascience 4 1ndash16


Chia J-M Song C Bradbury PJ Costich D de Leon N Doebley J Elshire RJ Gaut B Geller L Glaubitz JC et al (2012) Maize HapMap2identifies extant variation from a genome in flux Nat Genet 44 803ndash807


Choulet F Alberti A Theil S Glover N Barbe V Daron J Pingault L Sourdille P Couloux A Paux E et al (2014) Structural andfunctional partitioning of bread wheat chromosome 3B Science 345 1249721ndash1249721


Clark SE Williams RW Meyerowitz EM (1997) The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floralmeristem size in arabidopsis Cell 89 575ndash585


Crowell S Falcao AX Shah A Wilson Z Greenberg AJ McCouch SR (2014) High-Resolution Inflorescence Phenotyping Using a NovelImage-Analysis Pipeline PANorama Plant Physiol 165 479ndash495


Crowell S Korniliev P Falcatildeo A Ismail A Gregorio G Mezey J McCouch S (2016) Genome-wide association and high-resolutionphenotyping link Oryza sativa panicle traits to numerous trait-specific QTL clusters Nat Commun 7 10527


Danilevskaya ON Meng X Ananiev E V (2010) Concerted modification of flowering time and inflorescence architecture by ectopicexpression of TFL1-like genes in maize Plant Physiol 153 238ndash251


Eveland AL Goldshmidt A Pautler M Morohashi K Liseron-Monfils C Lewis MW Kumari S Hirag S Yang F Unger-Wallace E et al(2014) Regulatory modules controlling maize inflorescence architecture Genome Res 24 431ndash443


Fletcher JC Brand U Running MP Simon R Meyerowitz EM (1999) Signaling of Cell Fate Decisions by CLAVATA3 in ArabidopsisShoot Meristems Science 283 1911ndash1914


Gage JL Miller ND Spalding EP Kaeppler SM de Leon N (2017) TIPS a system for automated image-based phenotyping of maizetassels Plant Methods 13 21


Gage JL White MR Edwards J Kaeppler S de Leon N (2018) Selection signatures underlying dramatic male inflorescencetransformation during modern hybrid maize breeding bioRxiv


Hall H Ellis B (2012) Developmentally equivalent tissue sampling based on growth kinematic profiling of Arabidopsis inflorescencestems New Phytol 194 287ndash296



Hamblin MT Salas Fernandez MG Casa AM Mitchell SE Paterson AH Kresovich S (2005) Equilibrium Processes Cannot Explain HighLevels of Short- and Medium-Range Linkage Disequilibrium in the Domesticated Grass Sorghum bicolor Genetics 171 1247ndash1256


Hariprasanna K Rakshit S (2016) Economic Importance of Sorghum In S Rakshit Y-H Wang eds The Sorghum Genome SpringerInternational Publishing pp 1ndash25


Hart GE Schertz KF Peng Y Syed NH (2001) Genetic mapping of Sorghum bicolor (L) Moench QTLs that control variation in tilleringand other morphological characters Theor Appl Genet doi 101007s001220100582


Huang X Qian Q Liu Z Sun H He S Luo D Xia G Chu C Li J Fu X (2009) Natural variation at the DEP1 locus enhances grain yield inrice Nat Genet 41 494ndash497


Hudson RR Slatkin M Maddison WP (1992) Estimation of levels of gene flow from DNA sequence data Genetics 132 583ndash589Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Jeong S Trotochaud AE Clark SE (1999) The Arabidopsis CLAVATA2 Gene Encodes a Receptor-like Protein Required for the Stabilityof the CLAVATA1 Receptor-like Kinase Plant Cell 11 1925ndash1933


Kawahara Y de la Bastide M Hamilton JP Kanamori H McCombie WR Ouyang S Schwartz DC Tanaka T Wu J Zhou S et al (2013)Improvement of the Oryza sativa Nipponbare reference genome using next generation sequence and optical map data Rice 6 1ndash10


Kerstetter R a Laudencia-Chingcuanco D Smith LG Hake S (1997) Loss-of-function mutations in the maize homeobox gene knotted1are defective in shoot meristem maintenance Development 124 3045ndash54


Kusmec A Schnable PS (2018) FarmCPUpp Efficient large-scale genomewide association studies Plant Direct 2 e00053Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Kusmec A Srinivasan S Nettleton D Schnable PS (2017) Distinct genetic architectures for phenotype means and plasticities in Zeamays Nat Plants 3 715ndash723


L Heffner E Sorrells M Jannink J-L (2009) Genomic Selection for Crop Improvement Crop Sci 49Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Laux T Mayer KF Berger J Juumlrgens G (1996) The WUSCHEL gene is required for shoot and floral meristem integrity in ArabidopsisDevelopment 122 87ndash96


Li X Li X Fridman E Tesso TT Yu J (2015) Dissecting repulsion linkage in the dwarfing gene Dw3 region for sorghum plant heightprovides insights into heterosis Proc Natl Acad Sci 112 11823ndash11828


Li X Zhu C Yeh C-T Wu W Takacs EM Petsch KA Tian F Bai G Buckler ES Muehlbauer GJ et al (2012) Genic and nongeniccontributions to natural variation of quantitative traits in maize Genome Res 22 2436ndash2444


Liu X Huang M Fan B Buckler ES Zhang Z (2016) Iterative Usage of Fixed and Random Effect Models for Powerful and EfficientGenome-Wide Association Studies PLOS Genet 12 e1005767


Lunde C Hake S (2009) The interaction of knotted1 and thick tassel dwarf1 in vegetative and reproductive meristems of maizeGenetics 181 1693ndash1697

Pubmed Author and Title wwwplantphysiolorgon March 3 2020 - Published by Downloaded from Copyright copy 2018 American Society of Plant Biologists All rights reserved

Google Scholar Author Only Title Only Author and Title

Mackay TFC Stone EA Ayroles JF (2009) The genetics of quantitative traits challenges and prospects Nat Rev Genet 10 565ndash577Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Mayer KFX Schoof H Haecker A Lenhard M Juumlrgens G Laux T (1998) Role of WUSCHEL in regulating stem cell fate in theArabidopsis shoot meristem Cell 95 805ndash815


McSteen P (2006) Branching out the ramosa pathway and the evolution of grass inflorescence morphology Plant Cell 18 518ndash22Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Millan B Aquino A Diago MP Tardaguila J (2017) Image analysis-based modelling for flower number estimation in grapevine J SciFood Agric 97 784ndash792


Morris GP Ramu P Deshpande SP Hash CT Shah T Upadhyaya HD Riera-Lizarazu O Brown PJ Acharya CB Mitchell SE et al (2013)Population genomic and genome-wide association studies of agroclimatic traits in sorghum Proc Natl Acad Sci 110 453ndash458


Nagaraja Reddy R Madhusudhana R Murali Mohan S Chakravarthi DVN Mehtre SP Seetharama N Patil J V (2013) Mapping QTL forgrain yield and other agronomic traits in post-rainy sorghum [Sorghum bicolor (L) Moench] Theor Appl Genet 126 1921ndash1939


Paterson AH Bowers JE Bruggmann R Dubchak I Grimwood J Gundlach H Haberer G Hellsten U Mitros T Poliakov A et al (2009)The Sorghum bicolor genome and the diversification of grasses Nature 457 551ndash556


Pautler M Tanaka W Hirano HY Jackson D (2013) Grass meristems I Shoot apical meristem maintenance axillary meristemdeterminacy and the floral transition Plant Cell Physiol 54 302ndash312


Raj A Stephens M Pritchard JK (2014) FastSTRUCTURE Variational inference of population structure in large SNP data setsGenetics 197 573ndash589


Salas Fernandez MG Bao Y Tang L Schnable PS (2017) A High-Throughput Field-Based Phenotyping Technology for Tall BiomassCrops Plant Physiol 174 2008ndash2022


Schnable JC (2015) Genome Evolution in Maize From Genomes Back to Genes Annu Rev Plant Biol 66 329ndash343Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Schnable JC Freeling M Lyons E (2012) Genome-wide analysis of syntenic gene deletion in the grasses Genome Biol Evol 4 265ndash277Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Segura V Vilhjaacutelmsson BJ Platt A Korte A Seren Uuml Long Q Nordborg M (2012) An efficient multi-locus mixed-model approach forgenome-wide association studies in structured populations Nature Genetics 44 825ndash830


Srinivas G Satish K Madhusudhana R Nagaraja Reddy R Murali Mohan S Seetharama N (2009) Identification of quantitative trait locifor agronomically important traits and their association with genic-microsatellite markers in sorghum Theor Appl Genet 118 1439ndash1454


Stephens JC Miller FR Rosenow DT (1967) Conversion of Alien Sorghums to Early Combine Genotypes Crop Sci 7 396Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Taguchi-Shiobara F Yuan Z Hake S Jackson D (2001) The fasciated ear2 gene encodes a leucine-rich repeat receptor-like proteinthat regulates shoot meristem proliferation in maize Genes Dev 15 2755ndash2766



Tanaka W Pautler M Jackson D Hirano HY (2013) Grass meristems II Inflorescence architecture flower development and meristemfate Plant Cell Physiol 54 313ndash324


Thurber CS Ma JM Higgins RH Brown PJ (2013) Retrospective genomic analysis of sorghum adaptation to temperate-zone grainproduction Genome Biol 14 R68


Tsuda K Ito Y Sato Y Kurata N (2011) Positive Autoregulation of a KNOX Gene Is Essential for Shoot Apical Meristem Maintenance inRice Plant Cell 23 4368ndash4381


Vollbrecht E Reiser L Hake S (2000) Shoot meristem size is dependent on inbred background and presence of the maize homeoboxgene knotted1 Development 127 3161ndash3172


Vollbrecht E Springer PS Goh L Buckler IV ES Martienssen R (2005) Architecture of floral branch systems in maize and relatedgrasses Nature 436 1119ndash1126


Wallace JG Bradbury PJ Zhang N Gibon Y Stitt M Buckler ES (2014) Association Mapping across Numerous Traits Reveals Patternsof Functional Variation in Maize PLoS Genet doi 101371journalpgen1004845


Wang Q Tian F Pan Y Buckler ES Zhang Z (2014) A SUPER powerful method for genome wide association study PLoS Onedoi101371journalpone0107684


Wu X Li Y Shi Y Song Y Zhang D Li C Buckler ES Li Y Zhang Z Wang T (2016) Joint-linkage mapping and GWAS reveal extensivegenetic loci that regulate male inflorescence size in maize Plant Biotechnol J 14 1551ndash1562


Xu G Wang X Huang C Xu D Li D Tian J Chen Q Wang C Liang Y Wu Y et al (2017) Complex genetic architecture underlies maizetassel domestication New Phytol 214 852ndash864


Yang J Benyamin B McEvoy BP Gordon S Henders AK Nyholt DR Madden PA Heath AC Martin NG Montgomery GW et al (2010)Common SNPs explain a large proportion of heritability for human height Nat Genet 42 565ndash569


Yang J Lee SH Goddard ME Visscher PM (2011) GCTA A tool for genome-wide complex trait analysis Am J Hum Genet 88 76ndash82Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title

Youssef HM Eggert K Koppolu R Alqudah AM Poursarebani N Fazeli A Sakuma S Tagiri A Rutten T Govind G et al (2016) VRS2regulates hormone-mediated inflorescence patterning in barley Nat Genet 49 157ndash161


Yu J Pressoir G Briggs WH Vroh Bi I Yamasaki M Doebley JF McMullen MD Gaut BS Nielsen DM Holland JB et al (2006) A unifiedmixed-model method for association mapping that accounts for multiple levels of relatedness Nat Genet 38 203ndash208


Zhang D Kong W Robertson J Goff VH Epps E Kerr A Mills G Cromwell J Lugin Y Phillips C et al (2015) Genetic analysis ofinflorescence and plant height components in sorghum (Panicoidae) and comparative genetics with rice (Oryzoidae) BMC Plant Biol15 107


Zhang D Yuan Z (2014) Molecular Control of Grass Inflorescence Development Annu Rev Plant Biol 65 553ndash578Pubmed Author and TitleGoogle Scholar Author Only Title Only Author and Title wwwplantphysiolorgon March 3 2020 - Published by Downloaded from


Zhang Z Ersoz E Lai C-Q Todhunter RJ Tiwari HK Gore MA Bradbury PJ Yu J Arnett DK Ordovas JM et al (2010) Mixed linearmodel approach adapted for genome-wide association studies Nat Genet 42 355ndash360


Zhao S Gu J Zhao Y Hassan M Li Y Ding W (2015) A method for estimating spikelet number per panicle Integrating image analysisand a 5-point calibration model Sci Rep 5 16241


Zhao J Mantilla Perez MB Hu J Salas Fernandez MG (2016) Genome-Wide Association Study for Nine Plant Architecture Traits inSorghum Plant Genome 9 (2)1-14 doi 103835plantgenome2015060044



Parsed Citations

Reviewer PDF

Parsed Citations

2

Semi-automatedfeatureextractionfromRGBimagesforsorghumpanicle25 architectureGWAS26 Yan Zhou1 Srikant Srinivasan12 Seyed Vahid Mirnezami3 Aaron Kusmec1 Qi Fu45 Lakshmi 27 Attigala1 Maria G Salas Fernandez1 Baskar Ganapathysubramanian3 Patrick S Schnable14 28

1Department of Agronomy Iowa State University Ames Iowa 50011 USA 29

2Current Address School of Computing and Electrical Engineering Indian Institute of 30 Technology Mandi Kamand Himachal Pradesh 175005 India 31

3Department of Mechanical Engineering Iowa State University Ames 50011 Iowa USA 32

4College of Agronomy China Agricultural University 100083 Beijing China 33

5Current Address Molecular Breeding Lab BGI Research 518083 Shenzhen China 34 35

One-sentencesummary Image-based phenotyping facilitates the genetic dissection of 36 uni- and multi-dimensional traits of sorghum panicles 37

AuthorContributions38 PSS and BG conceived the original research plan and supervised the experiments and 39 data analyses PSS YZ and SS designed the experiments SS and SVM developed the 40 phenotype extraction pipeline and conducted phenotype extractions YZ conducted most 41 of the experiments and analyses QF assisted with collecting phenotypic data and image 42 processing AK assisted with GWAS and genetic analyses MGSF assisted with the 43 interpretation of results PSS LA MGSF and YZ wrote the article with contributions 44 from all the authors 45

Fundingsources46 This material is based upon work that was supported in part by the National Institute of 47 Food and Agriculture US Department of Agriculture (award no 2012-67009-19713) to 48 PSS and MGSF and colleagues and additional funding from ISUrsquos Plant Sciences Institute 49 to PSS MGSF and BG 50 Address correspondence to schnableiastateedu 51 52 53

Abstract 54


3




4



5


166


6

Results167



7




8




9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

3




4



5


166


6

Results167



7




8




9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

4



5


166


6

Results167



7




8




9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

5


166


6

Results167



7




8




9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

6

Results167



7




8




9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

7




8




9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

8




9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

9



10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

10



11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

11




12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

12



13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

13



14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

14



15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

15


353


16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

16



17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

17



18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

18





19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

19





20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

20



GenotypingData507


21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

21





22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

22






23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

23








24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

24



25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

25









Weietal2014




Brownetal2011




w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

26


33 Largerpanicle(LP






Brownetal2011

w

ww

plantphysiolorgon M


Dow

nloaded from

Copyright copy

2018 Am

erican Society of P

lant Biologists A

ll rights reserved

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

27

597

FigureLegends598









28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

28



29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

29

666 667 668 669

























































































































































Parsed Citations

Reviewer PDF

Parsed Citations
























































































































































Parsed Citations

Reviewer PDF

Parsed Citations

Download - 1 Short title: Semi-automated Sorghum Panicle Phenotyping 2 · 157 panicle phenotypes from a subset of the Sorghum Association Panel (SAP) (Casa et al., 158 2008) consisting of 272

Top Related