Computational Drug Discovery and Design

(backadmin) #1
computational expenses and to slowing down the performance
substantially.


  1. These values ofkandttMDwere calibrated for GAGs of length
    up to heptasaccharide [14]. An increase of k and a decrease of
    ttMDcould lead to the insufficient conformational sampling of
    the ligand during the targeted MD step. On the contrary, a
    decrease ofkand an elongation of the targeted MD step would
    lead to the scenario when many ligands are not properly tar-
    geted to the receptor but would be uniformly distributed on
    the sphere defined by a distance restraint, which would lead to
    the need of many more repetitions of the procedure to obtain a
    statistical ensemble of docking solutions suitable for further
    analysis. These values ofkandttMDfor significantly longer
    ligands, however, should be calibrated in terms of translational
    and rotational freedom by comparing their RMSD in an MD
    simulation with the RMSD expected from the targeting
    through the shortest possible path.

  2. In order to substantially decrease the invested computational
    time, it is practically convenient to remove all waters and
    counterions from the structure of the obtained complex, to
    add them again and to use the equilibration protocol described
    in Subheading2.2 (steps 5and 6 ) before the free MD step. In
    this case the minimum distance from solute atoms to the box
    boundaries could be increased to 8–9 A ̊ to avoid possible
    artifacts related to the insufficient treatment of system’s
    hydration.

  3. Some runs can end up with the ligand in an unbound state after
    the tMD step or the ligand can unbind during the free MD
    step. Depending on how many events like this are observed,
    100 repeated procedures could be enough or not. We recom-
    mend obtaining at least 80–90 bound docking poses for the
    further analysis. In case of longer GAG ligands this number can
    be increased to obtain higher statistical significance of the
    results.


Acknowledgments


This work was supported by National Science Center of Poland
(Narodowy Centrum Nauki, grant UMO-2016/21/P/ST4/
03995). This project received funding from the European Union’s
Horizon 2020 research and innovation programme under the
Marie Skłodowska-Curie grant agreement No. 665778.

Solvent Inclusion in Docking Glycosaminoglycans 453
Free download pdf