Premium
Enhancing the Reliability of GPCR Models by Accounting for Flexibility of Their Pro‐Containing Helices: the Case of the Human mAChR1 Receptor
Author(s) -
Pedretti Alessandro,
Mazzolari Angelica,
Ricci Chiara,
Vistoli Giulio
Publication year - 2015
Publication title -
molecular informatics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.481
H-Index - 68
eISSN - 1868-1751
pISSN - 1868-1743
DOI - 10.1002/minf.201400159
Subject(s) - g protein coupled receptor , computational biology , computer science , virtual screening , flexibility (engineering) , reliability (semiconductor) , template , biological system , metric (unit) , data mining , bioinformatics , biology , pharmacophore , mathematics , power (physics) , receptor , engineering , genetics , physics , statistics , operations management , quantum mechanics , programming language
To better investigate the GPCR structures, we have recently proposed to explore their flexibility by simulating the bending of their Pro‐containing TM helices so generating a set of models (the so‐called chimeras) which exhaustively combine the two conformations (bent and straight) of these helices. The primary objective of the study is to investigate whether such an approach can be exploited to enhance the reliability of the GPCR models generated by distant templates. The study was focused on the human mAChR1 receptor for which a presumably reliable model was generated using the congener mAChR3 as the template along with a second less reliable model based on the distant β2‐AR template. The second model was then utilized to produce the chimeras by combining the conformations of its Pro‐containing helices (i.e., TM4, TM5, TM6 and TM7 with 16 modeled chimeras). The reliability of such chimeras was assessed by virtual screening campaigns as evaluated using a novel skewness metric where they surpassed the predictive power of the more reliable mAChR1 model. Finally, the virtual screening campaigns emphasize the opportunity of synergistically combining the scores of more chimeras using a specially developed tool which generates highly predictive consensus functions by maximizing the corresponding enrichment factors.