Background: Gene set analysis based on Gene Ontology (GO) can be a promising method for the analysis of differential expression patterns. However, current studies that focus on individual GO terms have limited analytical power, because the complex structure of GO introduces strong dependencies among the terms, and some genes that are annotated to a GO term cannot be found by statistically significant enrichment. Results: We proposed a method for enriching clustered GO terms based on semantic similarity, namely cluster enrichment analysis based on GO (CeaGO), to extend the individual term analysis method. Using an Affymetrix HGU95aV2 chip dataset with simulated gene sets, we illustrated that CeaGO was sensitive enough to detect moderate expression changes. When compared to parent-based individual term analysis methods, the results showed that CeaGO may provide more accurate differentiation of gene expression results. When used with two acute leukemia (ALL and ALL/AML) microarray expression datasets, CeaGO correctly identified specifically enriched GO groups that were overlooked by other individual test methods. Conclusion: By applying CeaGO to both simulated and real microarray data, we showed that this approach could enhance the interpretation of microarray experiments. CeaGO is currently available at http://chgc.sh.cn/en/software/CeaGO/.

Over the last decade, emerging research methods, such as comparative genomic analysis and phylogenetic study, have yielded new insights into genotypes and phenotypes of closely related bacterial strains. Several findings have revealed that genomic structural variations (SVs), including gene gain/loss, gene duplication and genome rearrangement, can lead to different phenotypes among strains, and an investigation of genes affected by SVs may extend our knowledge of the relationships between SVs and phenotypes in microbes, especially in pathogenic bacteria. In this work, we introduce a ‘Genome Topology Network’ (GTN) method based on gene homology and gene locations to analyze genomic SVs and perform phylogenetic analysis. Furthermore, the concept of ‘unfixed ortholog’ has been proposed, whose members are affected by SVs in genome topology among close species. To improve the precision of 'unfixed ortholog' recognition, a strategy to detect annotation differences and complete gene annotation was applied. To assess the GTN method, a set of thirteen complete M. tuberculosis genomes was analyzed as a case study. GTNs with two different gene homology-assigning methods were built, the Clusters of Orthologous Groups (COG) method and the orthoMCL clustering method, and two phylogenetic trees were constructed accordingly, which may provide additional insights into whole genome-based phylogenetic analysis. We obtained 24 unfixable COG groups, of which most members were related to immunogenicity and drug resistance, such as PPE-repeat proteins (COG5651) and transcriptional regulator TetR gene family members (COG1309). The GTN method has been implemented in PERL and released on our website. The tool can be downloaded from http://homepage.fudan.edu.cn/zhouyan/gtn/ , and allows re-annotating the ‘lost’ genes among closely related genomes, analyzing genes affected by SVs, and performing phylogenetic analysis. With this tool, many immunogenic-related and drug resistance-related genes were found to be affected by SVs in M. tuberculosis genomes. We believe that the GTN method will be suitable for the exploration of genomic SVs in connection with biological features of bacterial strains, and that GTN-based phylogenetic analysis will provide additional insights into whole genome-based phylogenetic analysis.