Seite - 128 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

4. ConclusionandFutureWork ThisworkshowsthatFIGSapproachthroughmachinelearningalgorithmswereefficient infinding environmental signals for the characterization traitsmeasured as part of con- servation efforts doneby the ICARDAgenebank.Thehighpredictability ofmodels for all traitswas used later in predicting the non-evaluated accessions hold at ICARDA to assign probabilities for the characterization traits to accessions and stored in ICARDA genebank database and used as predictive characterization. This will assist genebank managersat ICARDAonreplyingmoreprecisely toseedrequests fromusers.Thiswork will be enhanced by usingmolecularmarkers techniques to finetune further FIGS ap- proach. References [1] Upadhyaya,H.D.,Ortiz,R.,Bramel, P.J. et al.,Development of a groundnut core collectionusing tax- onomical, geographical andmorphologicaldescriptors,GeneticResourcesandCropEvolution, (2003), https://doi.org/10.1023/A:1022945715628 [2] Upadhyaya,HDandBramel,PJandSingh,S,Developmentofachickpeacoresubsetusinggeographic distributionandquantitative traits,CropScience, (2001) [3] Gepts,P.,PlantGeneticResourcesConservationandUtilization,CropSci., (2006) [4] El Bouhssini,M. , Street, K. , Amri, A. ,Mackay,M. , Ogbonnaya, F. C., Omran, A. , Abdalla, O. , Baum,M. ,Dabbous,A. andRihawi, F., Sources of resistance in breadwheat toRussianwheat aphid (Diuraphis noxia) in Syria identified using the Focused Identification ofGermplasmStrategy (FIGS), PlantBreeding, (2011), doi:10.1111/j.1439-0523.2010.01814.x [5] KhazaeiH,StreetK,BariA,MackayM,StoddardFL,TheFIGS(FocusedIdentificationofGermplasm Strategy)Approach Identifies Traits Related toDrought Adaptation inVicia fabaGenetic Resources, PLoSONE, (2013), doi:10.1371/journal.pone.0063107 [6] ThormannI,Parra-QuijanoM,RubioTesoML,EndresenDTF,DiasS, IriondoJM,MaxtedN.,Predic- tive characterizationmethods for accessingandusingCWRdiversity, In:MaxtedN,DullooME,Ford- LloydBV, eds. Enhancing crop genepool use. Capturingwild relative and landrace diversity for crop improvement,Boston:CABI International, (2016) [7] Endresen,D.T. F.,K. Street,M.Mackay,A.Bari,A.Amri, E.DePauw,K.Nazari, andA.Yahyaoui, SourcesofResistance toStemRust (Ug99) inBreadWheatandDurumWheat IdentifiedUsingFocused IdentificationofGermplasmStrategy,CropSci., (2012), doi:10.2135/cropsci2011.08.0427 [8] Bari, A., Street, K., Mackay, M. et al., Focused identification of germplasm strategy (FIGS) de- tects wheat stem rust resistance linked to environmental variables, Genet Resour Crop Evol, (2012), https://doi.org/10.1007/s10722-011-9775-5 [9] BARI, A., AMRI, A., STREET, K.,MACKAY,M., DE PAUW, E., SANDERS, R., NAZARI, K., et al., Predicting resistance to stripe (yellow) rust (Puccinia striiformis) inwheat genetic resources using focused identificationofgermplasmstrategy,TheJournalofAgriculturalScience, (2014) [10] Noelle L. Anglin, Ahmed Amri, Zakaria Kehel, and Dave Ellis, A Case of Need: Linking Traits to GenebankAccessions,BiopreservationandBiobanking, (2018) [11] Fick,S.E. andR.J.Hijmans,Worldclim2:New1-kmspatial resolutionclimate surfaces forglobal land areas, International JournalofClimatology, (2017) [12] TitleP.O.,BemmelsJ.B.,ENVIREM:anexpandedsetofbioclimaticandtopographicvariablesincreases flexibilityand improvesperformanceofecologicalnichemodeling,Ecography, (2018) [13] S.B.Kotsiantis,Supervisedmachine learning:a reviewofclassification techniques, Informatica, (2007) [14] C.-W.Hsu,C.-C.Chang,C.-J. Lin,APracticalGuide toSupportVectorClassification,Department of ComputerScience,NationalTaiwanUniversity,Taipei,Taiwan, (2010) [15] Breiman,L. ,Randomforests,MachineLearning, (2001) [16] R.Rojas,NeuralNetwoks:ASystematic Introduction,Springer-Verlag,Berlin (1996) [17] A.Kołcz,N-tupleNetwork,CART,andBagging, inNeuralComputation, (2000) Z.Azoughetal. /PredictiveCharacterizationof ICARDAGenebankBarleyAccessions128