The inclusion of evolutionarily related genes as modules revealed a striking number of families with expression patterns associated with stemness. Duplication and subsequent modification of genes and their enhancers through evolution may have provided the control logic needed to diversify stem cells for populating new tissue systems distinct from ancestral counterparts. Indeed, the homolog modules consistently received higher S-MAP scores compared to the modules in a previous study that did not consider gene homology. The identification of stemness modules has obvious implications for both iPS cells and cancer therapeutics. Reprogramming candidates essential for stem cell properties are likely found among the stemness-on modules while stemness-off modules likely contain genes that have to be silent in stem cells. The SI scoring strategy, possibly adapted specifically to ESC/iPS cell signatures, could be used in the characterization of putative iPS lines. Similar strategies can be used to identify and target CSC within a heterogeneous cancer population and serve as valuable tools for fighting cancer progression. In addition, the results can be used to test for novel stem cell marker genes to improve the definition and isolation of stem cells. The S-MAP approach synergizes with large-scale efforts to systematically extract differentially expressed genes in many different conditions represented in current microarray gene expression repositories, such as Oncomine and GeneChaser. It is complementary to previous approaches to uncover the mechanisms of pluri- and multi-potency maintenance, such as the development of the PluriNet. In this study, we focused our analysis on modules with expression patterns ranging across a broad array of different stem cell types. However, S-MAP also revealed 152 homolog and 32 functional modules that were restricted to specific stem cell types; these modules should reveal tissue- or lineage-specific stem cell mechanisms. The method synergizes with large-scale efforts to systematically extract differentially expressed genes in many different conditions represented in current microarray gene expression repositories, such as Oncomine and GeneChaser. Finally, S-MAP can be used to test for inter-species stem cell expression Ginsenoside-F2 conservation, to characterize cancers from their normal counterparts, and is readily transferrable to other systems and diverse types of data. While the S-MAP procedure could score all collected modules, we defined a nonredundant set of modules to use for assessing the overall significance and distribution of scores. Allowing Albaspidin-AA redundancy could skew the results because of highly characterized pathways. For example, Gene Ontology contains a deep hierarchy with many sub- and sub-sub-categorizations under the cell cycle process because it has been studied extensively in yeast. A redundancy filter on the functional modules helped us avoid redundant annotations of highly characterized pathways and protein complexes from dominating the results. To filter out redundant functional modules, we first sorted the functional modules from smallest to largest and excluded any having a 25% or greater gene overlap with a smaller functional module. Functional modules with over half of the genes belonging to the same set of homologs were also excluded because any such module also reflects an evolutionary connection between the genes to a greater degree than the functional connection. The relatively high overlap cutoff between evolutionary and functional gene modules allows genes with multiple functional roles to be captured by different module types. The recurrence score used for S-MAP was motivated by the need to compensate for potential biases in combining heterogeneous results. There are obvious variations in the overlap and sizes of the SGLs and DGLs, which reflect the differences in the compared biological specimens, cell, and RNA isolation techniques, hybridization protocols, and statistical methods applied to identify differentially expressed clones.