"Many protein complexes prediction approaches are
based on the assumptions that (i) protein complexes have dense
protein-protein interactions (PPI) among their subunits, and (ii)
high functional similarity for the subunits. We suggest to
investigate those assumptions by studying the subunits’
interaction topology and sequences identity. Such
consideration can possibly provide better insights for our
understanding of protein complexes architecture. Two
topological parameters, density of protein-protein interaction
(PPI) and connectedness, are defined to test whether protein
complex are found in PPI dense region or not. The present
data indicated that interaction dense regions represent protein
complexes up to 20% cases. A rather large proportion of
protein complexes have a lower density of PPI, and
connectedness. It is conjectured that prediction approaches
based on the assumption that complexes are composed of
highly PPI dense regions, connectedness can predict a rather
limited numbers of the complexes. On the other hand, our
result indicates that protein complexes do composing of
subunits with higher sequence identity or similarity. "
Relation:
IEEE International conference on computer research and development 2011 (ICCRD 2011)
