Mısırda in vivo haploid katlama yöntemlerine karşı indirgeyici ve donör genotiplerin tepkilerinin araştırılması ve haploid tohum tespitinde kullanılan yöntemlerin karşılaştırılması
[ X ]
Tarih
2020
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Çanakkale Onsekiz Mart Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Bu tez çalışması mısır ıslahında kullanılan in vivo haploid katlama yöntemlerine karşı kullanılan materyallerin yanıtlarının irdelenmesi ve haploid tohum ayrımında kullanılan tekniklerin karşılaştırılması amacıyla yürütülmüştür. Çalışmada materyal olarak üç donör (DNR1, DRN4, DNR5) ile iki indirgeyici hat (CIM2GTAIL-P2, RWK-76xRWS) kullanılmıştır. Donör materyaller ve indirgeyici hatlar 2019 yılı Eylül ayında sera şartlarında KH0 tohumların elde edilmesi için yetiştirilmiştir. Serada yapılan tozlama işlemlerinin ardından donörlerden elde edilen tohumlardan haploid ve diploid olanlar göz ile sınıflandırılmıştır. Sınıflanan tohumlar kodlanmış ve sırası karışmayacak şekilde önce masaüstü tarayıcıda resimleri daha sonra da masaüstü NIR (Yakın Kızıl Ötesi) spektroskopi cihazında reflektans verileri toplanmıştır. KH0 tohumlar Zabirova ve diğerleri (1996) ve CIMMYT (Uluslararası Buğday ve Mısır Araştırma Merkezi) tarafından önerilen iki farklı yöntem ile kromozom katlama işlemine tabi tutulmuştur. Bu işlemin ardından fideler yeterli büyüklüğe ulaştığında 2020 yılı Mayıs ayında tarlaya şaşırtılmış ve bitki görünümü ile sap renklenmesi dikkate alınarak bitkilerin sınıflama durumları kayıt altına alınmıştır. Başlangıçta yapılan sınıflamadan elde edilen veriler ile son sınıflama verileri bağımlı değişken olarak kabul edilerek spektral veriler ve görüntü verileri (Mavi, Yeşil, Kırmızı bant) tahminleyici olarak kullanılmış ve sınıflama modelleri oluşturulumuştur. Oluşturulan modellerin sınıflama kapasiteleri gerçek sınıflama sonuçları ile kıyaslanarak değerlendirilmiştir. Araştırma bulgularına göre kullanılan donör ve indirgeyici hatların kromozom katlama yöntemlerine karşı tepkilerinin farklı olduğu görülmüştür. Haploid indirgenme oranları %9,2 ile %16,1 arasında değişim göstermiştir. CIMMYT yönteminde hem fideleme hem tarlaya şaşırtma aşamalarında bitki kayıplarının yüksek olduğu gözlenmiştir. Haploid tohum sınıflama amacıyla kullanılan teknikler bakımından da dikkate değer farklar olduğu belirlenmiştir. Yaygın olarak kullanılan göz ile ayırma yönteminin doğruluk oranı %87 olarak tespit edilir iken tohum görüntülerine ve spektral verileri dayalı oluşturulan modellerin doğruluk oranları sırasıyla %79 ve %83 olarak bulunmuştur. Bu çalışmanın sonuçları hem sepktral tekniklerin hem de görüntü işleme tekniğinin mısır ıslah programlarında in vivo maternal haploid tekniğinde sınıflama aracı olarak kullanılabileceğini ortaya koymuştur.
This thesis study has been conducted in order to examine the responses of the materials used against the in vivo haploid folding methods used in maize breeding, and also to compare the techniques used in haploid seed segregation. In this study work, three donors (DNR1, DRN4, DNR5) and two inducers line (CIM2GTAIL-P2, RWK-76xRWS) were used as materials. Donor materials and inducer lines have been grown under greenhouse conditions in September 2019 for the purpose of obtaining DH0 seeds. Haploid and diploid ones from seeds obtained from donors were classified by naked eye after the completion processes of pollination under greenhouse conditions. The classified seeds were coded and firstly, their pictures have been taken with the help of a desktop scanner and then reflectance data were collected on a benchtop NIR (Near Infrared Reflectance) spectroscopy device aim to not be mixed with each other. DH0 seeds have been subjected to chromosome folding using two different proposed methods by Zabirova et al (1997) and CIMMYT (International Maize and Wheat Improvement Center). After this process, when the seedlings reached a sufficient size, they were transplanted into the field in May 2020 and the classification of the plants has been recorded by considering the plant appearance and the colour of stalk. The data obtained from the initial classification and the final classification data were accepted as dependent variables, spectral data and image data (Blue, Green, Red band) were used for prediction and classification models generation. The classification capacities of the models have been evaluated by comparing them with the actual results of classification. According to the findings of current study, it has been observed that the reactions of the donor and used inducer lines were found different against the chromosome doubling methods. Haploid reduction rates shown a variation between 9.2% and 16.1%. In the CIMMYT method, plant losses have been observed high both in seedling as well as transplanting stages. It has also been observed that there were considerable differences in terms of techniques used for haploid seed classification. Accuracy rates of models created based on seed images and spectral data were found to be 79% and 83%, respectively, while the accuracy rate of widely used naked eye-separation method was determined as 87%. The overall results of this study revealed that both of the spectral techniques and image processing technique could be used as classification tools for in vivo maternal haploid technique in maize breeding programs.
This thesis study has been conducted in order to examine the responses of the materials used against the in vivo haploid folding methods used in maize breeding, and also to compare the techniques used in haploid seed segregation. In this study work, three donors (DNR1, DRN4, DNR5) and two inducers line (CIM2GTAIL-P2, RWK-76xRWS) were used as materials. Donor materials and inducer lines have been grown under greenhouse conditions in September 2019 for the purpose of obtaining DH0 seeds. Haploid and diploid ones from seeds obtained from donors were classified by naked eye after the completion processes of pollination under greenhouse conditions. The classified seeds were coded and firstly, their pictures have been taken with the help of a desktop scanner and then reflectance data were collected on a benchtop NIR (Near Infrared Reflectance) spectroscopy device aim to not be mixed with each other. DH0 seeds have been subjected to chromosome folding using two different proposed methods by Zabirova et al (1997) and CIMMYT (International Maize and Wheat Improvement Center). After this process, when the seedlings reached a sufficient size, they were transplanted into the field in May 2020 and the classification of the plants has been recorded by considering the plant appearance and the colour of stalk. The data obtained from the initial classification and the final classification data were accepted as dependent variables, spectral data and image data (Blue, Green, Red band) were used for prediction and classification models generation. The classification capacities of the models have been evaluated by comparing them with the actual results of classification. According to the findings of current study, it has been observed that the reactions of the donor and used inducer lines were found different against the chromosome doubling methods. Haploid reduction rates shown a variation between 9.2% and 16.1%. In the CIMMYT method, plant losses have been observed high both in seedling as well as transplanting stages. It has also been observed that there were considerable differences in terms of techniques used for haploid seed classification. Accuracy rates of models created based on seed images and spectral data were found to be 79% and 83%, respectively, while the accuracy rate of widely used naked eye-separation method was determined as 87%. The overall results of this study revealed that both of the spectral techniques and image processing technique could be used as classification tools for in vivo maternal haploid technique in maize breeding programs.
Açıklama
Fen Bilimleri Enstitüsü, Tarla Bitkileri Ana Bilim Dalı
Anahtar Kelimeler
Ziraat, Agriculture
