Files in this item



application/pdfNicola_Serao.pdf (4MB)
(no description provided)PDF


Title:Study of genetic by environmental interactions at transcriptomic and genomic levels
Author(s):Serao, Nicola
Director of Research:Rodriguez-Zas, Sandra L.
Doctoral Committee Chair(s):Rodriguez-Zas, Sandra L.
Doctoral Committee Member(s):Bollero, German A.; Beever, Jonathan E.; Shike, Daniel W.
Department / Program:Animal Sciences
Discipline:Animal Sciences
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Genome-wide association study (GWAS)
feed efficiency
gene expression
genetic by environment interaction
beef cattle
survival analysis
brain cancer
association study
single nucleotide polymorphisms (SNP)
residual feed intake (RFI)
Abstract:Many phenotypes in humans and animals are resulted by the interaction between genes and environment. For example, animals from distinct breeds may have different nutritional requirements, and thus, may respond in a particular way for each diet. In a similar fashion, genetic variability in the human population is responsible for the difference in drug efficiency on the treatment of several diseases. The development of high-throughput genomic and transcriptomics technologies has been providing rapid access to DNA and RNA information in humans and cattle. The access to this genetic characterization is a cornerstone step in understanding the complexity of characteristics within populations of these organisms. In humans, much of the attention on gene by environment interaction is brought for the understanding of several diseases. Glioblastoma multiforme (GBM), a grade IV cancer, is a complex multifactorial disorder that has swift and devastating consequences. With the objective of identifying general and clinical-dependent biomarkers for biomarkers for GBM, the expression pattern of 22,277 probes from 320 individuals diagnosed with GBM was evaluated. Three GBM time-to-event variables were considered: lifetime, overall and progression-free GBM survival. A novel analytical strategy was developed to identify general associations between the biomarkers and GBM, and associations that depend on cohort groups. A novel approach consisted of a five-step analytical strategy was developed in order to identify biomarkers for GBM. Results were further investigated through gene network inference, cross-validation and functional analyses. Through the novel analytical strategy, a total of 61, 47 and 60 gene expression profiles were simultaneously significantly associated (P-value < 0.05) with lifetime, overall, and progression-free survival, respectively. Many of the gene biomarkers associated had been previously reported with GBM (35, 24, and 35 genes, respectively) or with other cancer (10, 19, and 15 genes, respectively), and new associations were uncovered. Sixteen, four, and ten novel genes were associated with lifetime, overall and progression-free GBM survival, respectively. In addition, several genes were overrepresented among the three GBM events: PIK3R1, E2F3, AKR1C3, CSF1, JAG2, PLCG1, RPL37A, SOD2, TOPORS, HRAS, MDM2, CAMK2G, FSTL1, IL13RA1, MTAP and TP53. Of the genes associated with GBM on a cohort-manner, C2, EGFR, PRKCB, IGF2BP3, and GDF10 had gender-dependent associations; SOX10, RPS20, RAB31, and VAV3 had race-dependent associations; CHI3L1, PRKCB, POLR2D, and APOOL had therapy-dependent associations. These associations are the basis for individualized prognostic and gene-based therapies. The functional analyses identified many important biological processes and molecular functions associated with GBM, such as morphogenesis, cell cycle, aging, response to stimuli, and programmed cell death. In addition to the confirmation of biomarkers previously associated with GBM, novel general and cohort-dependent gene profiles were uncovered. These findings support the development of more accurate and personalized prognostic tools to improve the survival and quality of life of individuals afflicted by glioblastoma multiforme. In cattle, the genomic information may be accessed through the use of dense single nucleotide polymorphisms (SNP) arrays that cover the bovine genome. One of the main goals is the identification of SNPs that show associations with economically important traits for the beef cattle industry. The largest variable cost in the beef cattle industry comes from feed and may represent from 62 to 84% of the total costs. In this manner, the relationship between feed intake and production needs to be assessed. The genomic potential for feed efficiency was evaluated in a feedlot beef cattle population in two distinct studies. For both studies, 1,321 steers from five different genetic compositions fed five diets were genotyped using a 50K SNP panel. The data was divided in two independent data sets, in order to identify (training) and to test (validation) the associations. In the first study, general, breed- and diet-dependent associations for feed efficiency were identified using SNPs and haplotypes. The traits evaluated were: the two-step feed efficiency indicators residual feed intake (RFI), residual average daily gain (RADG), and residual intake gain (RIG), and two complementary one-step indicators of feed efficiency, efficiency of intake (EI) and efficiency of gain (EG). These two novel indicators were developed to account for the total variation removed in the one-step indicators. In addition, a multi-SNP model was developed to assess the predictive power of several SNPs. Network and functional analyses of genes associated with feed efficiency aided in the interpretation of the results. Thirty-one, 40, and 25 SNPs (P-value < 0.0001), and six, ten, and nine haplotypes (P-value < 0.001) were significantly associated with feed efficiency on a general, breed-dependent, and diet-dependent manners, respectively. The associations of 17 SNPs and 7 haplotypes were confirmed (P-value < 0.05) on the validation data set. Overlapping of 20 SNPs and six haplotype associations between RFI and EI, and five SNPs and one haplotype associations between RADG and EG, confirmed the complementary value of the one and two-step indicators. A total of 89 SNPs were included (P-value < 0.0001) in the multi-SNP models, and offered a precise prediction of the five feed efficiency indicators. Thirteen molecular functions and six biological processes were identified (P-value < 0.001) in the functional analysis, including ion channel activity, nucleotide binding, and passive transmembrane transporter activity. These Gene Ontology categories were overrepresented among the genes harboring SNPs associated with feed efficiency. The breed- and diet-dependent associations between SNPs and feed efficiency suggest that further refinement of variant panels require the consideration of the breed and management practices. To conclude, the unique genomic variants associated with the one- and two-step indicators suggest that both types of indicators offer complementary description of feed efficiency. In the second study, the feed efficiency components of average daily gain (ADG) and dry matter intake (DMI) adjusted for the maintenance requirements were used for SNP associations. Univariate and bivariate analyses were performed in order to assess feed efficiency in the training population. As in the first study, a multi-SNP model was developed, as well as functional and network analyses were performed. The bivariate model identified 11 significant associations (P-value < 0.0001), whereas the univariate analyses of ADG and DMI resulted in eight and nine associations, respectively. Of these, six SNPs were confirmed in the validation data set. The final multi-SNP model included seven, nine, and eight SNPs for the bivariate, and univariate ADG and DMI analyses, respectively. These models showed low drop in the model adequacy in the validation data set, amounting for 19.4, 11.68, and 7.21% compared to the training data set, respectively. Six Gene Ontology categories were (P-value < 0.001) identified for the SNPs associated (P-value < 0.001) in the bivariate model. These were all represented by molecular functions related to ion transport activity. The bivariate analysis of ADG and DMI helped in the identification of SNP that have beneficial associations with both components of feed efficiency and can be used for genome-enabled improvement of feed efficiency in feedlot beef cattle. These studies showed the importance of considering environmental factors, such as therapies and diets, interacting at transcriptomics and genomic levels in human and livestock research. These findings may have direct use in human health, through the development of gene-based therapies in GBM, and in beef cattle production, and by the incorporation of specific SNP panels in different production systems.
Issue Date:2013-02-03
Rights Information:Copyright 2012 Nicola Serao
Date Available in IDEALS:2013-02-03
Date Deposited:2012-12

This item appears in the following Collection(s)

Item Statistics