As we show below, this is a particularly acute problem for high frequency domains that are likely to occur near each other by chance, and very likely led to an overestimate of structural clustering. Conversely, in all such studies it is unclear how much information regarding gene clustering is missed due to the high levels of sequence divergence that INCB18424 JAK inhibitor arises between those ancestrally duplicated genes that have remained in proximity since the common ancestor of all vertebrates and even longer; detecting such arrangements requires the use of more sensitive approaches such as those involving Hidden Markov Models. In this present study our aim is to obtain new metrics and catalogues of clustered genes that both share structural features detected at the domain level and are highly significant in their proximity to one another, while allowing for sequence divergence to extend to the superfamily level. Several other previous studies have examined clusters of functionally related genes defined by their shared expression patterns, functional pathway enrichment or shared Gene Ontology annotations. These efforts attempted to distinguish whether such clusters arose by proximate gene duplications or by Paclitaxel company migration together at a common site by correcting for the presence of tandemly duplicated genes recognized by their sequence similarity. However, as we show below, a comprehensive description of genomic functional organization also requires considering clusters of ancestrally related genes that are difficult to recognize by sequence similarity methods alone or because they include significant numbers of interspaced, unrelated genes. Because duplications of genomic material provided a key evolutionary step in the origin of chromosomal arrangements, in this analysis we have applied the concept of paralogy, shared ancestry, to include genes that share a duplicated region recognizable as a common functional domain that arose as an domain duplication, as well as including genes that obviously share their entire sequence as a consequence of a prior, full gene duplication. Recent advancements in both the quality and quantity of gene annotation data have made it possible to assert gene paralogy by combining data from five sources of structural annotation data. We use the term paraclusters to represent clusters of paralogous genes that share sufficient structural similarity to imply a common ancestry, either in their entirety or in the possession of a common functional domain, and that are located in close proximity, where proximity is defined by a low probability that they would occur within the same neighborhood by chance. The long-term maintenance of these arrangements depends on the balance between possible purifying selection to maintain a functional advantage and random genomic rearrangements that promote disruption and dispersion.