| Summary: | 碩士 === 國立陽明大學 === 生物醫學資訊研究所 === 99 === With the popularization of high-throughput technology, enrichment tools have been rapidly developed for analysing large-scale ``omics'' data. However, most methods emphasize statistical significance rather then biological considerationand have difficulty assigning correct statistical significance to terms with few entities. It is therefore difficult for researchers to figure out accurate biological interpretation and assess the quality of Gene Ontology (GO) enrichment results.
In this study, we introduce a new functional enrichment analysis strategy. It integrates:
1)comparative genes/proteins quantization from experiments
2)the evidence code of GO annotation for quality control
3)the interaction relationship provided by STRING
to figure out the GO terms with accurate biological interpretation.
The output is expected to be precise to describe the experiments. In addition, we provide several output styles with graphic visualization. The PPI within terms, the DAG structure and gene similarity between terms are considered to cluster enriched GO terms.
Applying our strategy to the p53 +/- status expression dataset, the enriched term with the highest score is GO:0010640 (platelet-derived growth factor receptor signaling pathway, F3 gene, F7 gene), which is supported by literature. Since most of the top-ranked GO terms in the results are supported by previous study, we believe that the genes or proteins in the enriched terms have potential to be candidates for biomarker discovery or targets for experimantal design.
|
|---|
