Bioinformatics analysis of liver and adipose tissue microarray data from periparturient dairy cattle and development of a web-based ruminant specific microarray database

Bioinformatics;Systems Biology;Transcriptome profiling;Enrichment analysis;Bovine liver;Bovine adipose tissue;Transition cows

Shahzad, Khuram ; Drackley, James K., ; ,Hurley, Walter L. ; Loor ; Juan J.

关键词: Bioinformatics;    Systems Biology;    Transcriptome profiling;    Enrichment analysis;    Bovine liver;    Bovine adipose tissue;    Transition cows;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/42472/Khuram_Shahzad.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

To-date bovine liver and adipose tissues have been investigated through several molecular techniques. However, for the most part molecular work in these tissues has been conducted using a reductionist approach discussing the individual genes, proteins or pathways. This approach does not necessarily evaluate the overall biological effect of a particular condition (e.g. treatment, physiological state) on a pathway. In some instances these tissues have been investigated at the transcriptomic and proteomic levels using several bioinformatics techniques such as gene enrichment analysis or over-represented approach (ORA).The ORA analysis is widely-used in scientific community for functional analysis of high-throughput data such as microarray or RNA-sequencing. It is performed to associate gene sets with their relevant biological pathways or functions using statistical methodologies. However, this approach has several limitations including inability to compare time series experiments or multiple treatments within the same experiment. In order to overcome these limitations, the Dynamic Impact Approach (DIA) has been developed by our group. This approach takes the concept of dynamism into consideration using proportion of differentially expressed genes compared to the genes present in array, statistically significant p-values and fold change values. This approach provides a holistic view of dynamic changes along time series experiments by means of impact and flux. The impact and flux take into account the overall biological perturbation and its direction (e.g., either induced or inhibited) to interpret the biological functions. For our time series microarray datasets, we employed both the DIA and enrichment analysis approaches together to infer the biological meanings. The current studies are mainly focused on the systematic evaluation of the tissues during the periparturient period, i.e. the last three weeks through the first three weeks around parturition. The results indicated a significant amount of biological adaptation in the liver and adipose tissue during pregnancy into early lactation. The large amount of data generation and curation needs a way to be systematically stored, retrieved and analyzed using bioinformatics techniques. For this purpose, we have laid the foundation of a ruminant specific microarray database (Ruminant Physiome) for the data mining and representation. The objectives of this research were, 1) to evaluate the metabolic responses in the liver of dairy cows that were overfed or restricted-fed dietary energy prepartum, 2) to evaluate the metabolic changes and physiological adaptations in the bovine adipose tissue occurring during pregnancy into early lactation, and 3) to develop a ruminant specific ‘omics’ website to provide public access to the expression profiling datasets along with their functional results.