Emerging methods in protein co-evolution

2015-03-08

Emerging methods in protein co-evolution

Extracted Annotations (3/8/2015, 11:05:57 AM)

"Darwin himself initiated the study of co-evolution, and his observation on the relationship between the size of orchids' corollae and the length of the proboscis of pol- linators led him to predict successfully the existence of a new species that was able to suck from the large spur of Darwin's orchid." (Demaria et al 2014:722)

"term co-evolution is usually attrib- uted to Ehrlich 3 ," (Demaria et al 2014:722)

"'reciprocal evolutionary change in interacting species' 4 ." (Demaria et al 2014:722)

"Methods for detecting co-evolution at the protein level often mine phylogenetic trees that have been built for many protein families using entire protein sequences" (Demaria et al 2014:722)

"prediction of contacts in protein structures, sites of specific protein interactions and the predictions of protein interaction partners at the genomic scale." (Demaria et al 2014:723)

"the trees can be converted to distance matrices to quantify the tree similarity" (Demaria et al 2014:723)

"Mutual information has been also used to detect co-varying positions" (Demaria et al 2014:724)

"Protein contact prediction" (Demaria et al 2014:724)

"Specificity-determining positions (SDPs) are groups of positions that coordinately mutate in the context of subfamily divergence. T" (Demaria et al 2014:725)

"Statistical coupling analysis (SCA)-like approaches have successfully been used to explore the implication of networks of co-evolving residues in protein folding 50 and allosteric communication 51,53 (see the lower panel of the figure)," (Demaria et al 2014:725)

"Monte Carlo algorithm An algorithm based on simulated repeated random sampling to obtain approximate solutions to complex mathematical and statistical problems." (Demaria et al 2014:727)

"residue entropy" (Demaria et al 2014:727)

"multiple correspondence analysis 48 (MCA), which is conceptually equivalent to PCA but is better suited to dealing with categorical data, such as amino acid identities" (Demaria et al 2014:729)

here (note on p.730)



"So far, we have separated co-evolution-based methods into two classes: those that use the MSA of a single protein family to detect (intra-protein) residue co-evolution and those that use MSAs of two families to detect inter-protein co-evolution, including the search for the interaction region of proteins known to interact and the search in complete genomes for the interaction partners of a given protein" (Demaria et al 2014:731)

"For example, this can avoid undesired crosstalk with recently appeared signalling pathways, or it can maintain the interaction with a rapidly evolving protein." (Demaria et al 2014:732)