Session: COMP Poster Session:

The history and philosophy of chemical modeling started in 1968 in England and Israel higher education institutions. The first computer for modeling called Golem-1 was home built at Weizmann Institute in Israel. The computer language utilized back in the late 1960s was fortran, a language for math and science purposes. Current chemical modeling computer codes utilize C,C++,Fortran or Python computer languages. The first chemical modeling project in 1968 was done with a series of nalkanes and cycloalkanes molecules and performing conformational,vibrational spectra and enthalpies calculations on computers. The method utilized was molecular mechanics based rather than quantum mechanics. The workup of the data used least squares optimization to look at best fit between theoretical calculations and experimental value. The second major chemical modeling project back in 1969 was comparing the theoretical molecular mechanics based structure to x ray crystal structures of lysozyme and myoglobin proteins. The molecular mechanics field consisted of bond lengths,bond angles,dihedral angles,non bonded interactions and fifth term for restraint purposes to compensate for the differences between theoretical and experimental x ray structures. The math for the dihedral angles and non bonded interactions is also more complex in current molecular mechanic functions and another factor is the new polarization term and models. Suprisingly in 1969 the theoretical derived structures only vary slightly from x ray structures,root mean square deviation was 0.22 angstroms for lysozyme and root mean square deviation was 0.086 angstroms for myoglobin. Its still common in simulations today that molecular mechanics calculations have faster computing times than quantum mechanics calculations with more complex math functions. Size extensivity in quantum chemistry is the total energy scaling with increasing system size. A more recent quantum chemistry method can even do calculations on a large protein. One important point about both quantum mechanics and molecular mechanics methods is that researchers studying sets of homologous molecules is more common practice because the methods aren't suitable for every molecule. For example,new quantum mechanic methods have been found to do calculations on free radicals,the so called open shell molecules. Covalent and non covalent interactions are more accurately described by quantum mechanics rather than molecular mechanics methods.

Michael Renier

chemistry consultant

Appearances