Computational protein modelling based on limited sets of constraints

Paper in proceeding, 2016

Molecular dynamics simulations, often combined with sim- ulated annealing, are commonly used when calculating structural mod- els of proteins, e.g. based on NMR experiments. However, one is often faced with limited and sometimes insufficient information for determin- ing a well-resolved 3D structure. In addition, the type of data available for different proteins may vary: ranges for torsion angles, distance ap- proximations, relative orientation of different molecular parts etc. We are using whatever structural information is around, together with a dynamic programming approach for searching the space of feasible con- formations to rapidly determine 3D structures that are consistent with the input constraints. Time-efficiency is important for good sampling of the conformational space but also to replace expensive, complex and time consuming experiments. Our approach benefits from having both high level and low level descriptions of conformational features and con- straints, and the possibility to infer new constraints from those that are given.