Computational Approaches in the Design of Synthetic Receptors - Designing Receptors for the Next Generation of Biosensors - page 153

etc., in calculations aimed at prediction of optimal template-monomer pairs [ 75 ].
The imprinting factor was predicted by artificial neural networks as a function of
the calculated molecular and mobile phase descriptors. The quantum chemical
descriptors were computed using Gaussian 03W. The model confirmed that the
stronger template-functional monomer interactions lead to higher imprinting
factors.
Most experiments are influenced by a variety of factors, and screening is often
the first step in efficient assessment of which factors are important in influencing the
desired outcome of the system under study [ 183 , 184 ]. Factors that depend on
monitoring the effects of changing one factor at a time on a response are extremely
time consuming producing erroneous optimums in experiments [ 183 ].
Chemometrics often combined with artificial network simulations [ 75 ] can help
significantly in optimization and design of experiments for simultaneous changing
factors to predict optimum conditions thus reducing the number of experiments
necessary [ 183 , 184 ]. While chemometrics offers several advantages, complex
numerical solutions that are generated by chemometrics approaches could often
be misinterpreted unless a proper procedure is implemented. Since problem solving
involves multivariate analysis, the interpreter needs to be fairly skilled in terms of
analysis and interpretation. This approach also requires a number of experiments to
be made, and as such cannot be performed entirely in silico.
3 Conclusion
Various methods such as MM, MD, QC methods, and QM have been clearly
shown to be extremely useful for the design of MIPs. Each of these methods
should take account of the various complex physical and chemical processes
taking place during the formation of monomer-template complexes, as well as
processes involved in polymer formation. QC methods in the recent past have
been heavily used because of their low computational cost and their ability to
provide relatively accurate estimations of the electronic structures of molecules
pertaining to non-covalent interactions in pre-polymerization mixtures. Generally
in the rational design of MIPs, it is important to note that although formation of
monomer-template complexes in solution can be relatively easily modeled, the
challenge still lies in modeling different stages of polymerization. The exact
mechanism of events related to the incorporation of monomers into the polymer
network and their effects on MIP recognition properties need to be more fully
understood. This would lead to a greater understanding of binding events and lead
to better design of MIPs. Future improvements in computational protocols could
also throw light onto the reasons for the discrepancies observed between the
predicted and experimental results of the performance of polymers where the
recognition process involves hydrophobic interactions. Current and future compu-
tational approaches would certainly help in the design of artificial receptors and
generate knowledge that could help to build a better understanding of molecular