Dataset Information

Evaluating gene set enrichment analysis via a hybrid data model.

ABSTRACT: Gene set enrichment analysis (GSA) methods have been widely adopted by biological labs to analyze data and generate hypotheses for validation. Most of the existing comparison studies focus on whether the existing GSA methods can produce accurate P-values; however, practitioners are often more concerned with the correct gene-set ranking generated by the methods. The ranking performance is closely related to two critical goals associated with GSA methods: the ability to reveal biological themes and ensuring reproducibility, especially for small-sample studies. We have conducted a comprehensive simulation study focusing on the ranking performance of seven representative GSA methods. We overcome the limitation on the availability of real data sets by creating hybrid data models from existing large data sets. To build the data model, we pick a master gene from the data set to form the ground truth and artificially generate the phenotype labels. Multiple hybrid data models can be constructed from one data set and multiple data sets of smaller sizes can be generated by resampling the original data set. This approach enables us to generate a large batch of data sets to check the ranking performance of GSA methods. Our simulation study reveals that for the proposed data model, the Q2 type GSA methods have in general better performance than other GSA methods and the global test has the most robust results. The properties of a data set play a critical role in the performance. For the data sets with highly connected genes, all GSA methods suffer significantly in performance.

SUBMITTER: Hua J

PROVIDER: S-EPMC3929260 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Evaluating gene set enrichment analysis via a hybrid data model.

Hua Jianping J Bittner Michael L ML Dougherty Edward R ER

Cancer informatics 20140212 Suppl 1

Gene set enrichment analysis (GSA) methods have been widely adopted by biological labs to analyze data and generate hypotheses for validation. Most of the existing comparison studies focus on whether the existing GSA methods can produce accurate P-values; however, practitioners are often more concerned with the correct gene-set ranking generated by the methods. The ranking performance is closely related to two critical goals associated with GSA methods: the ability to reveal biological themes an ...[more]

PMID: 24558298

Similar Datasets

Project description:BackgroundMethods for gene-class testing, such as Gene Set Enrichment Analysis (GSEA), incorporate biological knowledge into the analysis and interpretation of microarray data by comparing gene expression patterns to pathways, systems and emergent phenotypes. However, to use GSEA to its full capability with non-mammalian model organisms, a microarray platform must be annotated with human gene symbols. Doing so enables the ability to relate a model organism's gene expression, in response to a given treatment, to potential human health consequences of that treatment. We enhanced the annotation of a microarray platform from a non-mammalian model organism, and then used the GSEA approach in a reanalysis of a study examining the biological significance of acute and chronic methylmercury exposure on liver tissue of fathead minnow (Pimephales promelas). Using GSEA, we tested the hypothesis that fathead livers, in response to methylmercury exposure, would exhibit gene expression patterns similar to diseased human livers.ResultsWe describe an enhanced annotation of the fathead minnow microarray platform with human gene symbols. This resource is now compatible with the GSEA approach for gene-class testing. We confirmed that GSEA, using this enhanced microarray platform, is able to recover results consistent with a previous analysis of fathead minnow exposure to methylmercury using standard analytical approaches. Using GSEA to compare fathead gene expression profiles to human phenotypes, we also found that fathead methylmercury-treated livers exhibited expression profiles that are homologous to human systems & pathways and results in damage that is similar to those of human liver damage associated with hepatocellular carcinoma and hepatitis B.ConclusionsThis study describes a powerful resource for enabling the use of non-mammalian model organisms in the study of human health significance. Results of microarray gene expression studies involving fathead minnow, typically used for aquatic ecological toxicology studies, can now be used to generate hypotheses regarding consequences of contaminants and other stressors on humans. The same approach can be used with other model organisms with microarray platforms annotated in a similar manner.

Dataset Information

Evaluating gene set enrichment analysis via a hybrid data model.

Publications

Evaluating gene set enrichment analysis via a hybrid data model.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets