Chemistry Reference
In-Depth Information
compare and score stored fragments to identify suitable hits. Use of more powerful
machines can help address both of these issues, and is likely to lead to the
development of more ambitious and reliable tools in the future. The ability of
algorithms to predict suitable hits might be improved by the incorporation of
available binding site information, and by the use of machine learning techniques to
help characterize the origins of fragment activity. Such approaches are becoming
more common in the related field of the
de novo
fragment drug design [25].
As computational methods for bioisosteric prediction become more accepted and
reliable, their application to lead optimization problems will become increasingly
widespread. The ultimate measure of the benefit brought by each method will be
its ability to help chemists to make effective structural modifications quickly, and
to offer an increased diversity of chemical substituents by making novel
predictions that can complement the safer selection of known transformations that
are currently in use [5].
SCAFFOLD HOPPING
Scaffold hopping is a technique closely related to the prediction of bioisosterism.
Both scaffold hopping method and the bioisosteric replacement are used to
improve synthetic accessibility, potency and drug-like properties of an active
compound, as well as joining a new chemical space. The main idea of this
approach is to create new chemical entities by keeping a sufficient similarity with
the originalbioactive compound in order to maintain the biological activity after
changing the core of the compound and moving it to a new chemical space [3].
Scaffold hopping can be defined as a method used to discover new chemical
classes by replacing a part (scaffold, the core of the molecule) of a known
compound preserving the remaining chemical groups under the assumption that
they are important to biological activity [26].
Therefore, the challenge of scaffold hopping is to identify molecular cores which
will form the mainstay of new families of compounds with similar biological
activity to a known reference molecule. This goal is reached when through the
search of scaffolds, the relative group orientations that interact directly with the
target binding site are kept functional [5]. This prevents not only problems with
