Abstract
AbstractFor decades, scientists and engineers have been working
to predict protein interactions, and network topology methods
have emerged as extensively studied techniques. Recently,
approaches based on AlphaFold2 intelligence, exploiting 3D
molecular structural information, have been proposed for protein
interaction prediction, they are promising as potential
alternatives to traditional laboratory experiments, and their
design and performance evaluation is compelling.Here, we
introduce a new concept of intelligence termed Network Shape
Intelligence (NSI). NSI is modelled via network automata rules
which minimize external links in local communities according to a
brain-inspired principle, as it draws upon the local topology and
plasticity rationales initially devised in brain network science
and then extended to any complex network. We show that by using
only local network information and without the need for training,
these network automata designed for modelling and predicting
network connectivity can outperform AlphaFold2 intelligence in
vanilla protein interactions prediction. We find that the set of
interactions mispredicted by AlphaFold2 predominantly consists of
proteins whose amino acids exhibit higher probability of being
associated with intrinsically disordered regions. Finally, we
suggest that the future advancements in AlphaFold intelligence
could integrate principles of NSI to further enhance the
modelling and structural prediction of protein interactions.
Users
Please
log in to take part in the discussion (add own reviews or comments).
