My interests in computers stared with web design, and
computer graphics. In the last few years I have been
interested in molecular modeling. Molecular modeling is an
essential method used to help predict the main structural
features of a protein. More than often, from computer models,
it is possible to derive useful functional information about
the proteins in concerned.

Homology modeling is a technique for building a 3D
model of a protein of known sequence but unknown
structure, using known related structures as templates. In
this technique, two requirements need to be met; first the
target sequence (sequence with unknown structure) must
have significant similarity to another sequence with a
known 3D structure. Second, it must be possible to build
an alignment. The higher the sequence identity to the
template, the higher the accuracy of the model. This is
under the assumption that small changes in the protein
sequence usually result in a small change in its 3D
structure. There is a great deal of structural data to
support this supposition (Aloy et al. 2003), but it is
expected that the alignment falls in the >30% identity.
Although it is accepted that two proteins are homologous
if they have <30% identity, some proteins may still be
homologous even if they have very little sequence identity.
The means for detecting their similarity other than sequence
identity can be used such as threading techniques. If no
template is found, a model can be created by first
principles or with the used of peptide databases.