Aroop Sircar - Community-wide assessment of protein-interface modeling suggests improvements to design methodology

Document created by Aroop Sircar on Aug 22, 2014
Version 1Show Document
  • View in full screen mode

  Publication Details (including relevant citation   information):

  Journal of Molecular Biology, 414(2), 289-302 (2011)


  The CAPRI (Critical Assessment of Predicted Interactions) and   CASP (Critical Assessment of protein Structure Prediction)   experiments have demonstrated the power of community-wide tests   of methodology in assessing the current state of the art and   spurring progress in the very challenging areas of protein   docking and structure prediction. We sought to bring the power of   community-wide experiments to bear on a very challenging protein   design problem that provides a complementary but equally   fundamental test of current understanding of protein-binding   thermodynamics. We have generated a number of designed   protein–protein interfaces with very favorable computed binding   energies but which do not appear to be formed in experiments,   suggesting that there may be important physical chemistry missing   in the energy calculations. A total of 28 research groups took up   the challenge of determining what is missing: we provided   structures of 87 designed complexes and 120 naturally occurring   complexes and asked participants to identify energetic   contributions and/or structural features that distinguish between   the two sets. The community found that electrostatics and   solvation terms partially distinguish the designs from the   natural complexes, largely due to the nonpolar character of the   designed interactions. Beyond this polarity difference, the   community found that the designed binding surfaces were, on   average, structurally less embedded in the designed monomers,   suggesting that backbone conformational rigidity at the designed   surface is important for realization of the designed function.   These results can be used to improve computational design   strategies, but there is still much to be learned; for example,   one designed complex, which does form in experiments, was   classified by all metrics as a nonbinder

  Address (URL):