Accurate Biomolecular Structures by the Nano-LEGO Approach: Pick the Bricks and Build Your Geometry

The determination of accurate equilibrium molecular structures plays a fundamental role for understanding many physical-chemical properties of molecules, ranging from the precise evaluation of the electronic structure to the analysis of the role played by dynamical and environmental effects in tuning their overall behavior. For small semi-rigid systems in the gas phase, state-of-the-art quantum chemical computations rival the most sophisticated experimental (from, for example, high-resolution spectroscopy) results. For larger molecules, more effective computational approaches must be devised. To this end, we have further enlarged the compilation of available semiexperimental (SE) equilibrium structures, now covering the most important fragments containing H, B, C, N, O, F, P, S, and CI atoms collected in the new SE100 database. Next, comparison with geometries optimized by methods rooted in the density functional theory showed that the already remarkable results delivered by PW6B95 and, especially, rev-DSDPBEP86 functionals can be further improved by a linear regression (LR) approach. Use of template fragments (taken from the SE100 library) together with LR estimates for the missing interfragment parameters paves the route toward accurate structures of large molecules, as witnessed by the very small deviations between computed and experimental rotational constants. The whole approach has been implemented in a user-friendly tool, termed nano-LEGO, and applied to a number of demanding case studies.

Automated summary

Accurately determining equilibrium molecular structures is crucial for understanding molecular properties, with quantum chemical computations providing precise results for small molecules and more effective approaches needed for larger ones. The SE100 database now includes important fragments containing various atoms. Linear regression can improve geometries optimized by PW6B95 and rev-DSDPBEP86 functionals, paving the way for accurate structures of large molecules with small deviations between computed and experimental results. The nano-LEGO tool has been developed to implement this approach and applied to challenging case studies.