Chemistry Reference
In-Depth Information
low sequence identity, the correct alignment is difficult. The schemes use score
pairs of all aligned residues. For example, alignment of pairs of identical/similar
residues gets higher scores than dissimilar pairs. Scoring schemes includes
chemical similarity, observed substitution scoring and genetic codes.
During sequence alignment there is structural alignment of equivalent residues.
This includes active site regions and secondary structure elements. Gaps
(insertions and deletions often present in loop regions) can be determined in both
target and template structures. More efficient multiple alignment of multiple
structures can also be performed whereas various sequences are compared at once.
In the third step, regarding the modeling of structurally conserved regions, we
note the following. The homology approach is based on the fact that there are
nearly identical families (in their 3d structures) that corresponds to regions in the
proteins (inner cores). Here we can find differences in peptide chain topology
which could yield significant effects on protein structure. After alignment,
modeling of similar regions in the target can be done.
The core of the model can be built from equivalent aligned residues of the
templates using conserved secondary structure elements present in the template
structures and selecting sections from the template structures (based on an
estimation of the likelihood of similarity of local environments for individual
amino acids). As the number of structural templates increase, the assignment of
structural conserved regions becomes more important. There should be a careful
examination (superposition) to identify the more relevant regions. This is
important to validate the sequence alignment.
In the fourth step regarding loop modeling, we note that in these regions occurs
the most important structural differences. There are typically gaps in the
alignment between model and template sequences. In the model, these gaps are
related to deletions. In in the template sequence these gaps are related to
insertions. Although not straightforward to predict, these changes can be limited
to loop and turn regions.
Energy-based and knowledge-based are major procedures for loop modeling
protocols. In the energy-based method an energy function is used to classify the
