Whereas databases of experimentally detected protein complexes continue to serve the community well, they are inherently incomplete - especially when it comes to dynamic combinatorial complexes - and thus can only partially explain all the relevant interplay. The latter studies only considered a very limited complexome and took a simplistic view at differential abundances across cellular states. Approaches dealing with such interdependencies and the limitedness of gene products have been attempted by stochastic simulations with according computational effort and by linear optimization on fixed sets of reference complexes. However, such simplified models lack a ruleset addressing how proteins that are expressed in low amounts and that are shared between different binding partners may limit complex formation. Besides, the topic was examined by integrating expression data with known protein complexes and annotated pathways. Guided by static compilations of protein interactions, the correlation of gene expression or protein abundance between putative interaction partners was used as a proxy to study their collective behavior. Nowadays a plethora of data on gene expression and an increasing amount of data on proteome abundances enable to also approach the dynamics of the condition-specific complexome by computational methods. More so, direct quantitative measures are limited to a definite protein space and only cover pairwise complexation. Quantitative profiling of the complete complexome in a condition-specific way is currently not feasible in a high-throughput fashion. Whereas the experimental detection of protein complexes is generally speaking a mature field, it is still time-consuming and subject to high false-discovery rates. Such multiprotein complexes may be either clearly defined modules of interaction partners that represent permanently assembled molecular machines or combinatorial formations of transient interaction partners in a dynamic interplay. Instead, proteins frequently collide with other gene products in the crowded environment of the cell, they may selectively bind to other proteins driven by physical interactions, they may dynamically assemble into complexes in a well-coordinated manner and accomplish their tasks cooperatively. Ĭellular function is a team effort because proteins rarely perform their biochemical tasks all alone. A platform-independent Java binary, a user guide with example data and the source code are freely available at. ConclusionsĬompleXChange allows to analyze deregulation of the protein complexome on a whole-genome scale by integrating a plethora of input data that is already available. Furthermore we showed that deregulated complexes identified by the tool potentially harbor significant yet unused information content. We demonstrated that our new method is robust against false-positive detection and reports deregulated complexomes that can only be partially explained by differential analysis of individual protein-coding genes. The practical usability of the method was assessed in the context of transcription factor complexes in human monocyte and lymphoblastoid samples. We observed for simulated data that results obtained by our complex abundance estimation algorithm were in better agreement with the ground truth and physicochemically more reasonable compared to previous efforts that used linear programming while running in a fraction of the time. Here, we present a pipeline for differential analysis of protein complexes based on predicted or manually assigned complexes and inferred complex abundances, which can be easily applied on a whole-genome scale. This deficiency impedes the applicability of the powerful tool of differential analysis in the realm of macromolecular complexes. Consequently, there exist large amounts of transcriptomic data and an increasing amount of data on proteome abundance, but quantitative knowledge on complexomes is missing. Although a considerable number of proteins operate as multiprotein complexes and not on their own, organism-wide studies so far are only able to quantify individual proteins or protein-coding genes in a condition-specific manner for a sizeable number of samples, but not their assemblies.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |