DNA microarray has become a powerful tool in genetics, molecular biology, and biomedical research. DNA microarray can be used for measuring the genotypes, structural changes, and gene expressions of human genomes. Detection and characterization of multilevel, high-throughput microarray genomic data pose new challenges to statistical pattern recognition and machine learning research. In this dissertation, we propose novel computational methods for analyzing DNA copy number changes and learning the trees of phenotypes using DNA microarray data. DNA copy number change is an important form of structural variations in human genomes. The copy number signals measured by high-density DNA microarrays usually have low signal-to-noise ratios and complex patterns due to inhomogeneous composition of tissue samples. We propose a robust detection method for extracting copy number changes in a single signal profile and consensus copy number changes in the signal profiles of a population. We adapt a solution-path algorithm to efficiently solve the optimization problems associated with the proposed method. We tested the proposed method on both simulation and real CGH and SNP microarray datasets, and observed competitively improved performance as compared to several widely-adopted copy number change detection methods. We also propose a chromosome instability measure to summarize the extracted copy number changes for assessing chromosomal instabilities of tumor genomes. The proposed measure demonstrates distinct patterns between different subtypes of ovarian serous carcinomas and normal samples. Among active research on complex human diseases using genomic data, little effort and progress have been made in discovering the relational structural information embedded in the molecular data. We propose two stability analysis based methods to learn stable and highly resolved trees of phenotypes using microarray gene expression data of heterogeneous diseases. In the first method, we use a hierarchical, divisive visualization approach to explore the tree of phenotypes and a leave-one-out cross validation to select stable tree structures. In the second method, we propose a node bandwidth constraint to construct stable trees that can balance the descriptive power and reproducibility of tree structures. Using a top-down merging procedure, we modify the binary tree structures learned by hierarchical group clustering methods to achieve a given node bandwidth. We use a bootstrap based stability analysis to select stable tree structures under different node bandwidth constraints. The experimental results on two microarray gene expression datasets of human diseases show that the proposed methods can discover stable trees of phenotypes that reveal the relationships between multiple diseases with biological plausibility. / Ph. D.

DNA copy number change is an important form of structural variation in human genome. Somatic copy number alterations (CNAs) can cause over expression of oncogenes and loss of tumor suppressor genes in tumorigenesis. Recent development of SNP array technology has facilitated studies on copy number changes at a genome-wide scale, with high resolution. Quantitative analysis of somatic CNAs on genes has found broad applications in cancer research. Most tumors exhibit genomic instability at chromosome scale as a result of dynamically accumulated genomic mutations during the course of tumor progression. Such higher level cancer genomic characteristics cannot be effectively captured by the analysis of individual genes. We introduced two definitions of chromosome instability (CIN) index to mathematically and quantitatively characterize genome-wide genomic instability. The proposed CIN indices are derived from detected CNAs using circular binary segmentation and wavelet transform, which calculates a score based on both the amplitude and frequency of the copy number changes. We generated CIN indices on ovarian cancer subtypes' copy number data and used them as features to train a SVM classifier. The experimental results show promising and high classification accuracy estimated through cross-validations. Additional survival analysis is constructed on the extracted CIN scores from TCGA ovarian cancer dataset and showed considerable correlation between CIN scores and various events and severity in ovarian cancer development. Currently our methods have been integrated into G-DOC. We expect these newly defined CINs to be predictors in tumors subtype diagnosis and to be a useful tool in cancer research. / Master of Science

Doctor of Philosophy / Department of Statistics / Haiyan Wang / High throughput screening technologies have generated a huge amount of biological data in the last ten years. With the easy availability of array technology, researchers started to investigate biological mechanisms using experiments with more sophisticated designs that pose novel challenges to statistical analysis. We provide theory for robust statistical tests in three flexible models. In the first model, we consider the hypothesis testing problems when there are a large number of variables observed repeatedly over time. A potential application is in tumor genomics where an array comparative genome hybridization (aCGH) study will be used to detect progressive DNA copy number changes in tumor development. In the second model, we consider hypothesis testing theory in a longitudinal microarray study when there are multiple treatments or experimental conditions. The tests developed can be used to detect treatment effects for a large group of genes and discover genes that respond to treatment over time. In the third model, we address a hypothesis testing problem that could arise when array data from different sources are to be integrated. We perform statistical tests by assuming a nested design. In all models, robust test statistics were constructed based on moment methods allowing unbalanced design and arbitrary heteroscedasticity. The limiting distributions were derived under the nonclassical setting when the number of probes is large. The test statistics are not targeted at a single probe. Instead, we are interested in testing for a selected set of probes simultaneously. Simulation studies were carried out to compare the proposed methods with some traditional tests using linear mixed-effects models and generalized estimating equations. Interesting results obtained with the proposed theory in two cancer genomic studies suggest that the new methods are promising for a wide range of biological applications with longitudinal arrays.

High-throughput genomic datasets obtained from microarray or sequencing studies have revolutionized the field of molecular biology over the last decade. The complexity of these new technologies also poses new challenges to statisticians to separate biological relevant information from technical noise. Two methods are introduced that address important issues with normalization of array comparative genomic hybridization (aCGH) microarrays and the analysis of RNA sequencing (RNA-Seq) studies. Many studies investigating copy number aberrations at the DNA level for cancer and genetic studies use comparative genomic hybridization (CGH) on oligo arrays. However, aCGH data often suffer from low signal to noise ratios resulting in poor resolution of fine features. Bilke et al. showed that the commonly used running average noise reduction strategy performs poorly when errors are dominated by systematic components. A method called pcaCGH is proposed that significantly reduces noise using a non-parametric regression on technical covariates of probes to estimate systematic bias. Then a robust principal components analysis (PCA) estimates any remaining systematic bias not explained by technical covariates used in the preceding regression. The proposed algorithm is demonstrated on two CGH datasets measuring the NCI-60 cell lines utilizing NimbleGen and Agilent microarrays. The method achieves a nominal error variance reduction of 60%-65% as well as an 2-fold increase in signal to noise ratio on average, resulting in more detailed copy number estimates. Furthermore, correlations of signal intensity ratios of NimbleGen and Agilent arrays are increased by 40% on average, indicating a significant improvement in agreement between the technologies. A second algorithm called gamSeq is introduced to test for differential gene expression in RNA sequencing studies. Limitations of existing methods are outlined and the proposed algorithm is compared to these existing algorithms. Simulation studies and real data are used to show that gamSeq improves upon existing methods with regards to type I error control while maintaining similar or better power for a range of sample sizes for RNA-Seq studies. Furthermore, the proposed method is applied to detect differential 3' UTR usage.

Genomic prediction of single-crosses is a promising tool in maize breeding, increasing genetics gains and reducing selection time. A strategy that can increase accuracy is applying multiple-trait genomic prediction using selection indices, which take into account the performance under optimal and stress conditions. Moreover, factors such as dominance, structural variants, and population structure can influence the accuracy of estimates of genomic breeding values (GEBV). Therefore, the objectives were to apply genomic prediction (i) including multi-trait models, (ii) incorporating dominance deviation and copy number variation effects, and (iii) controlling population structure in maize hybrids. Hence, we used two maize datasets (HELIX and USP), consisting of 452 and 906 maize single-crosses. The traits evaluated were grain yield, plant and ear height, stay green, and four selection indices. From multi-trait GBLUP and GK, using the combination of selection indices in MTGP is a viable alternative, increasing the selective accuracy. Furthermore, our results suggest that the best approach is predicting hybrids including dominance deviation, mainly for complex traits. We also observed including copy number variation effects seems to be suitable, due to the increase of prediction accuracies and reduction of model bias. On the other hand, adding four different sets of population structure as fixed covariates to GBLUP did not improve the prediction accuracy for grain yield and plant height. However, using nonmetric multidimensional scaling dimensions and fineSTRUCTURE group clustering increased reliability of the GEBV for GY and PH, respectively. / Predição genômica de híbridos simples é uma promissora ferramenta no melhoramento de milho, pois permite aumentar os ganhos genéticos por unidade de tempo, principalmente por reduzir o tempo de seleção. Uma estratégia que pode aumentar a acurácia das predições genômicas é realizar esta para múltiplos caracteres considerando os mesmos simultâneamente, ou utilizar índices de seleção, os quais captam a performance dos genótipos tanto em condições ótimas como em condições de estresse. Além disso, fatores como dominância, variantes estruturais, e estruturação populacional podem influenciar a acurácia de estimativas dos valores genéticos genômicos (VGG). Portanto, os objetivos foram aplicar predição genômica em híbridos de milho (i) incluindo modelos multi-trait, (ii) incorporando desvios de dominância e efeitos da variação no número de cópias, e (iii) controlando a estruturação populacional. Para isto, dois conjuntos de milho (HELIX e USP) foram utilizados, consistindo de 452 e 906 híbridos simples. Os caracteres avaliados foram produtividade de grãos, altura de planta e espiga, senescência, e quatro índices de seleção. A partir das análises multi-trait dos modelos GBLUP e GK, pôde-se concluir que a combinação dos índices é uma alternativa viável, aumentando a acurácia seletiva. Além disso, os resultados sugerem que o melhor método é a predição de híbridos incluindo desvios de dominância, principalmente para caracteres complexos. Observou-se também que incluir efeitos relacionados a variação no número de cópias indica ser adequado, devido ao aumento da acurácia e redução do viés nos modelos de predição genômica. Por outro lado, a acurácia de predição não aumentou quando se adicionou quatro diferentes conjuntos de estruturação como covariáveis fixas no modelo GBLUP. No entanto, usando o escalonamento multidimensional não métrico e o agrupamento do fineSTRUCTURE aumentaram a confiabilidade de estimação do VGG para produtividade de grãos e altura de plantas, respectivamente.

O transtorno do espectro autista (ASD) é uma doença clinica e geneticamente heterogênea, com mecanismo etiológico ainda pouco conhecido. Assim, os principais objetivos deste trabalho foram descrever as características clínicas e genéticas de pacientes brasileiros com ASD, bem como determinar o risco de recorrência e a herdabilidade. Verificamos que a maioria das comorbidades avaliadas tem prevalência similar àquelas anteriormente descritas. A hipotonia exibiu maior prevalência no sexo feminino. A ausência de fala apresentou prevalência significativamente maior no grupo de pacientes com comorbidades, sendo que a gravidade da fala foi positivamente correlacionada com a presença das crises convulsivas. A herdabilidade estimada foi de 76% e o risco de recorrência ~5%. As alterações citogenéticas e os casos positivos para a Síndrome do X-Frágil explicaram cerca de 8% dos casos de ASD da nossa amostra. As CNVs nas regiões estudadas foram detectadas em 2,7% da amostra. Nós verificamos que há penetrância incompleta do ASD para as regiões. O estudo mais detalhado dos dois casos de duplicação da região 15q13.3, envolvendo somente o gene CHRNA7, mostrou que um dos pacientes (F5240) exibiu uma segunda CNV, possivelmente patogênica. A análise in silico sugeriu que genes que interagem diretamente com o CHRNA7 podem conter mutações patogênicas e, juntamente com a duplicação do 15q13.3, possivelmente estão envolvidos na etiologia do ASD. Este estudo mostrou que é necessário fazer uma ampla caracterização genética dos pacientes, para possibilitar o estudo dos possíveis mecanismos moleculares envolvidos na causa do ASD / Autism Spectrum Disorder (ASD) is a clinically and genetically heterogeneous disease and its etiological mechanisms are still poorly understood. The main objectives of this study were to describe the clinical and genetic features of Brazilian patients with ASD, and to determine the recurrence risk and heritability. Great part of the comorbidities assessed here had comparable prevalence to those of previous works. The hypotonia was significantly prevalent in the female sex. Absent speech was significantly more frequent in patients with comorbidities, and severity of speech problems was positively correlated with presence of seizures. Heritability was estimated as 76% and the recurrence risk as approximately 5%. Cytogenetic alterations and positive results for Fragile X Syndrome explain about 8% of the ASD etiology of our sample. The CNVs at the chromosomal regions 15q11-q13, 16p11.2 and 22q13 were present in 2.7% of the sample. Incomplete penetrance of ASD was observed for the 16p and 15q regions. Further investigation of the two cases with duplication of the region 15q13.3, involving only the CHRNA7 gene, revealed that one of them (F5240) exhibited a second possible pathogenic CNV. In silico analysis suggested that genes interacting directly with the CHRNA7 could harbor pathogenic mutations and, together with the duplication at 15q13.3, could be involved in the ASD etiology. This study showed the necessity of a broad genetic characterization of patients with ASD, to enable the elucidation of possible molecular mechanisms related to ASD etiology

Os transtornos do espectro autista (TEA) são condições neuropsiquiátricas caracterizadas por padrões comportamentais estereotipados, ausência ou limitação de comunicação verbal e de interação social recíproca. Diversos estudos têm mostrado que esses transtornos possuem etiologia genética complexa e heterogênea, o que dificulta a identificação dos fatores causais. Estima-se que cerca de 70% dos casos de TEA são idiopáticos. Portanto, com o objetivo de identificar mecanismos etiológicos associados aos TEA, utilizamos as seguintes estratégias: customização de uma lâmina de microarray CGH que possibilite a detecção não só de grandes CNVs, mas também de alterações menores do que 10 kbp, em exons e regiões UTR de genes potencialmente candidatos; a comparação entre os tipos de rearranjos detectados em pacientes sindrômicos e em não sindrômicos e, ainda, a investigação mais detalhada de uma família com indivíduos portadores de transtorno autista e síndrome de Asperger. Foram avaliados 103 portadores de TEA não sindrômicos e 18 sindrômicos, sendo as taxas de detecção de alterações potencialmente patogênicas, respectivamente, de 11,6% e 38,9%. Dentre as alterações detectadas 44,4% são menores do que 10 Kbp. Portanto, a estratégia de usar uma lâmina customizada, com alta densidade de sondas complementares aos exons e regiões não codificantes de genes potencialmente envolvidos na etiologia dos TEA, capaz de detectar tanto alterações grandes quanto pequenas, parece ser relevante na tentativa de elucidar o maior número de casos possíveis e melhor compreender esses transtornos. Além disso, essa lâmina também pode ser utilizada como uma ferramenta para auxiliar o diagnóstico clínico e o aconselhamento genético com um custo mais acessível em comparação a outras comerciais ou ao sequenciamento de última geração. Cerca de 33,3% das CNVs observadas afetam região UTR, sugerindo que mutações nessas regiões podem explicar uma proporção significativa dos casos nos quais não são detectadas alterações através de outros testes genômicos, visto que a maioria desses ainda não exploram adequadamente regiões não codificantes. Entre os pacientes autistas não sindrômicos verificou-se que a maioria dos genes afetados por CNVs estão envolvidos em duas principais funções biológicas - sinapses glutamatérgicas e orientação axonal, sugerindo que TEA não sindrômico pode ser causado por disfunção em genes diferentes envolvidos em processos fisiológicos comuns. Diferente do que observamos entre pacientes não sindrômicos, detectamos mais de uma alteração em um mesmo indivíduo ou alterações que englobam mais de um gene entre os pacientes sindrômicos, reforçando o modelo oligogênico para alguns casos de TEA. Por fim, os dados obtidos no estudo da família com portadores de síndrome de Asperger e transtorno autista sugere que a gravidade do quadro clínico possa estar relacionada ao número de mutações e possivelmente por duas mutações diferentes em ambos os alelos de um mesmo gene. Nossos resultados, além de apoiar o envolvimento dos genes MDGA2, FHIT, HTR2A, SHANK2, GRIA3, ZNF778, PRKCα, CDH15, DIAPH3, GCH1, GRM5, MARK1, SLC17A6, IMMP2L, BZRAP1, SYNGAP1, ANK3, MAP1A, GABRR2 e LAMC3 nos TEA também sugere novos genes candidatos: LRRC7, LRRIQ3, CADPS1, NUFIP, SEMA3A, SNAP29, MBD2, GAD2, DGKH e PARD3 / The autism spectrum disorders (ASD) are neuropsychiatric conditions typically characterized by social deficits, communication difficulties, stereotyped or repetitive behaviors and interests. Several studies have shown that these disorders have a complex and heterogeneous genetic etiology, which makes difficult to identify the causal factors. Approximately 70% of cases are idiopathic. In order to identify etiological mechanisms associated with ASD, we have used the following strategies: customized a microarray CGH platform that allows detection not only of large CNVs, but also alterations smaller than 10 kbp in exons and UTR regions of potential candidate genes, the comparison between the types of rearrangements detected in syndromic and non-syndromic patients and further, more detailed investigation of a family segregating both autistic disorder and Asperger syndrome. We evaluated 103 nonsyndromic and 18 syndromic patients by the custom-designed array and the detection rate of possibly pathogenic alterations were, respectively, 11.6% and 38.9%. Among these CNVs, 44.4% are smaller than 10 kbp. Therefore, the strategy of using a custom-designed array, enriched with probes targeted to genes potentially involved in the ASD etiology and able to detect both large and small CNVs, seems to be relevant in an attempt to elucidate the largest number of cases and to better understand these disorders. Furthermore, this platform can also be used as a tool to support the clinical diagnosis and genetic counseling with a more affordable cost compared to conventional other or next-generation sequencing. Approximately 33.3% of the observed CNVs affect UTR region, suggesting that mutations in non-coding regions might explain a significant proportion of ASD cases negative for most genomic screenings, which still do not explore adequately these regions. Among nonsyndromic autistic patients we found that most of the genes affected by CNVs are involved in two main biological functions - glutamatergic synapses and axonal guidance, suggesting that nonsyndromic ASD can be caused by dysfunction in different genes of a few common physiological processes. In contrast to our findings in nonsyndromic patients, we detected more than one alteration in a single individual or alterations that involve more than one gene among the syndromic patients, reinforcing the oligogenic model for some cases of ASD. Finally, the data obtained in the study of the family segregating both Asperger syndrome and autistic disorder suggests that the severity of ASD seems to be modulated by the number of hits and possibly by hits in both alleles of the same gene. Our results support the involvement of genes MDGA2, FHIT, HTR2A, SHANK2, GRIA3, ZNF778, PRKCα, CDH15, DIAPH3, GCH1, GRM5, MARK1, SLC17A6, IMMP2L, BZRAP1, SYNGAP1, ANK3, MAP1A, GABRR2 and LAMC3 in ASD etiology and also suggests new candidates: LRRC7, LRRIQ3, CADPS1, NUFIP, SEMA3A, SNAP29, MBD2, GAD2, DGKH and PARD3

Le cancer de l’ovaire est le cancer gynécologique avec la plus grande mortalité due à un diagnostique tardif au stade de maladie extensive péritonéale. Malgré les progrès de la chirurgie radicale et de la chimiothérapie les récurrences abdominales demeurent la cause la plus fréquente de mortalité. Il existe peu d’études de la maladie métastatique péritonéale. Notre hypothèse de travail est que les différences entre la maladie métastatique et la tumeur primaire sont primordiales dans la survenue d’une maladie résiduelle ou récurrente. Nous avons utilisé une approche exhaustive comprenant des études du transcriptome, des variations du nombre de copie (VNC) et des sequençages des exomes pour caractériser les différences entre lésions primaires, métastases péritonéales et métastases lymphatiques.Résultats: Notre étude démontre que les VNC varient de façon significative entre la tumeur primaire et la métastase peritonéale. Les différences d’expressions géniques bien que mineures permettent de retrouver les voies de signalisation primordiales pour le développement des métastases. Le séquençage des exomes montre très peu de différences en terme de polymorphisme. Par ailleurs la majorité des polymorphismes présents dans les métastases se retrouvent à une faible fréquence dans la tumeur primaire de façon concordante avec la théorie clonale. Conclusion: L’ensemble des résultats montre la possibilité d’une origine clonale de la maladie métastatique des cancers de l’ovaire comportant la majorité des anomalies au niveau des variations du nombre de copie. L’intégration de ces données permettrait d’optimiser les thérapeutiques ciblées. / Ovarian cancer is the most deadly gynecological cancer. The high rate of mortality is due to the large tumor burden with extensive metastatic lesion of the abdominal cavity. There are few studies on genetic alterations and their consequences in peritoneal metastatic tumors when compared to their matched ovarian primary tumors. Our hypothesis is that differences between the metastatic and primary lesions might be the cause of residual disease and, most importantly may have a role in post-chemotherapeutic recurrences. Methods: We conducted integrated genomics analysis on matched primary and metastatic tumors from 9 patients. In the papers presented here we analyze genome-wide Copy Number Variations (CNVs) using SNP Arrays targeting peritoneal metastasis differences, Gene expression differences using Microarrays also targeting peritoneal metastasis differences, and for some patients, Single Nucleotide Polymorphisms (SNPs) in genes through Exome sequencing.Results: Here we show that CNVs vary significantly between primary and metastatic tumors and include genes that have been considered potential chemotherapeutic targets based on primary tumor only data. Gene expression differences, while minor, showed highly statistically significant enrichment of genes in ovarian cancer critical pathways. In agreement with findings in other cancers, exome sequencing data revealed very few SNP differences of which most metastasis enriched SNPs were present at very low levels in the primary tumor. The results presented here should allow better design of therapies to target residual ovarian cancer disease.

Atualmente a procura por produtos sustentáveis têm-se mostrado cada vez mais frequente e promissora. Em espécies de importância comercial, procura-se obter a maior produtividade possível dentro de um curto espaço de tempo aliado à preservação do meio ambiente. Dentro disso, a transformação genética de plantas se mostra uma alternativa atrativa para a geração de variedades de cana-de-açúcar que gerem produtos de maneira mais eficaz. O sucesso da transformação genética está diretamente associada a cultura de tecidos de plantas que precisa ser adequada a cada genótipo e situação de cultivo, sendo a luminosidade um dos principais fatores para a produção de plantas vigorosas. Outro fator importante é a seleção das plantas transgênicas, que precisam ser submetidas a uma quantidade de agente seletivo suficiente para identificar as plantas modificadas geneticamente. Em cana-de-açúcar, a identificação de plantas transgênicas por PCR e a definição do número de cópias é um procedimento de difícil execução e muito oneroso. Isto se dá pois no processo transformação via biolística, a inserção de genes é aleatória, produzindo plantas com variados números de cópias. Em consideração a estes fatores envolvidos na eficiência de obtenção de plantas transgênicas de cana-de-açúcar, os objetivos deste trabalho foram o aperfeiçoamento do protocolo de cultura de tecidos, transformação genética da variedade SP803280 com os genes xth, AtDdm1, como também, definir genes de referência para a quantificação do número de cópias dos genes xth e AtDdm1 inseridos na variedade SP803280 e do gene neo na RB835089, análise de ploidia e tamanho de genoma dos eventos transgênicos comparado com as plantas controle. No estudo a respeito da melhor qualidade de luz durante o cultivo in vitro na fase de regeneração de plantas, tem-se que a luz branca e a junção das luzes LED e branca se mostraram melhores para regeneração e desenvolvimento das plantas enquanto que para plântulas, as luzes LED e branca separadamente foram mais efetivas no crescimento. Para a seleção das plantas uma concentração de geneticina entre 40 e 50 mgL-1 é recomendada. As taxas de sucesso nas transformações genéticas para o gene xth variaram entre 2,5 a 18,3% dependendo do experimento e para AtDdm1 foi de 2,2% em um bombardeamento.Não houveram alterações de ploidia e tamanho do genoma nos transgênicos das duas variedades em relação à planta selvagem. Os genes p4h e prr foram identificados como os melhores para a quantificação relativa de genes inseridos por PCR em tempo real na variedade SP803280 enquanto que para a RB835089 aprt e prr se mostraram mais eficazes. A análise do número de cópias inseridas em eventos transgênicos por PCR em tempo real foi possível através das duas metodologias de cálculo testados por este trabalho, com resultados que concordam com uma tendência nesta determinação de maneira simples e rápida. / Currently the demand for sustainable products has been shown to be frequent and promising. In species of commercial importance, there is an effort to obtain the highest possible productivity in a short time, along with the environment preservation. In this context, genetic transformation of plants appears as an attractive alternative for the development of sugarcane varieties able to generate products in a more effective way. The genetic transformation success is directly associated to plant tissue culture that requires specific condition for each genotype and cultivation process, in which, luminosity is one of the main factors that determines the production of vigorous plants. Another important factor is the selection of transgenic plants, that occur by exposing plants to a sufficient amount of selective agent in order to identify only genetic modified plants. In sugarcane, identification of transgenic plants by PCR and the definition of copy numbers is a difficult procedure to implement and usually is very costly. It is because in the process of genetic transformation by biolistic, the insertion of genes occurs randomly and also produce plants with varied copy numbers. In consideration of these factors directly involved in the efficiency to obtain sugarcane transgenic plants the objectives of this study were the improvement of a tissue culture protocol, the genetic transformation of the variety SP803280 with xth and AtDdm1 genes. Also, the studies include the definition of reference genes for determining the number of copies inserted of xth and AtDdm1 genes into the variety SP803280 and neo gene in RB835089, ploidy analysis and genome size of the transgenic events compared to control plants. In the study related to the best light quality for in vitro plant regeneration, white light and the combination of LED and white lights proved to be better for plants regeneration and development while for seedlings, LED and white light separately were more effective for growth. In order to obtain selection of transgenic plants, geneticin concentration between 40 and 50 mg L-1 is recommended. Success rates in xth genetic transformation ranged from 2.5 to 18.3% depending on the experiment, and for AtDdm1 was only 2.2% in just one biolistic bombardment. There were no changes in ploidy and genome size in transgenic events related to their wild type plant. The genes p4h and prr were defined to be the best for determining the copy number of transgenic events by real time PCR in SP803280 variety, while for RB835089, the genes aprt and prr were the most effective. The analysis of the number of inserted copies was possible using the two calculation methodologies tested by this work, with results that agree with a tendency in a simple and fast quantification methodology.

No description available.