According NU7441 concentration to functional classification of the Gene Ontology (GO) project [30], we selected several highly expressed genes within five different categories, including membrane receptors, TFs, growth factors and cytokines, chemokines, and signal-transduction molecules, in either dNK, cNK, or pNK cells. By integrating the data generated from the genomic profiling with information from published reports and bioinformatic databases

(e.g., STRING, Gene Network Central, Transcriptional Regulatory Element Database), we were able to determine that the genes highly expressed in NK cells formed a complex network, which was analyzed and visualized using the network analysis tool GeneMANIA [31] and the visualization software Cytoscape [32] (Fig. 1). Additionally, by combining these data with information available from published reports, bioinformatic databases and network analysis tools (such as STRING [33]), we were able to predict putative target genes of the selected TFs and finally describe the transcriptional regulatory networks of NK cells (Fig. 2). TFs including Ikaros, PU.1, Ets-1, Nfil3, Id2, T-bet,

and Eomes are key regulators that have a major effect on NK-cell fate, differentiation, and function. The target genes for all TFs examined in [60] were identified or predicted by searching click here published reports and online bioinformatic databases, including STRING [33], GeneMANIA [31], and TRED [34] (Fig. 3). The interaction network was visualized by Cytoscape software [32] (Fig. 3). In addition to Cytoscape, other visualization software including 3Dscape [35], Circos [36], and Gephi [37] are also available to integrate, analyze, and visualize the network data,

complex systems, dynamics, and hierarchical graphs. Overall, we think that integrating different analysis methods takes full advantage of what can be learned from the enormous amount of data generated from gene expression profiles. Many databases, software, and online tools are available Low-density-lipoprotein receptor kinase and useful for searching and predicting the function of gene sets and particular genes of interest. Moreover, we provide here a list of the databases, software, and online tools useful for this endeavor and include information on how the network biology tools and integrative informatics can be applied to large microarray datasets (Fig. 3 and Table 3). Finally, we illustrate how this strategy can be successfully applied to a large genomic expression profile dataset in our own studies in order to make further investigations into NK-cell biology. NK-cell subpopulations have a remarkable degree of repertoire and functional diversity. In humans, these diverse subpopulations include tolerant, cytotoxic, and regulatory NK cells [38].