Thu 11 Feb. 2010, 10:30

Title: Rational prediction, analysis and annotation of protein-protein interactions
Speaker: Igor Jurišica, University of Toronto

While high-throughput methods can help to identify large numbers
of protein and genetic interactions, and to interrogate gene
regulation, signalling, and metabolic networks and pathways,
their overlap is usually low. Even when combined with curated and
interologous interactions, this only covers about 13% of the
estimated human interactome. We need to accelerate and focus
these studies, and make optimal choices for the specific methods
to be used in order to characterize high priority cancer-related
targets. At present, approximately 40% of cancer signature genes
do not have protein-protein interaction information.

Many of these interactions can be predicted with high confidence
using data mining approach - fpClass algorithm. As a first step,
we consider UniProt with 74,946 proteins, comprising 36,899
proteins with non-redundant sequences and 23,013 protein
fragments. Using fpClass we have predicted ~2.6 billion pairs
with diverse probability of interaction among all 74,946
proteins. Second, we use text mining to increase the number of
papers supporting curated, high-throughput and predicted
interactions. Besides providing more confidence, this information
will provide context . potentially identifying multiple methods
used to detect a given PPIs, identifying region of the protein
used to detect PPI, in what tissue and under what conditions. In
turn, this may be useful to plan follow-up validation
experiments. Third, we use NAViGaTOR, a powerful graphing
application for the 2D and 3D visualization of biological
networks. NAViGaTOR includes a rich suite of visual mark-up tools
for manual and automated annotation, fast and scalable layout
algorithms, and OpenGL hardware acceleration to facilitate
scalable visualization of large graphs, combined with graph
theory algorithms to analyze topological features in networks.
Publication-quality images can be rendered through SVG graphics
export, and many community-developed data formats support data
import/export (PSI-XML, BioPax, and GML). NAViGaTOR's ability to
handle larger datasets is facilitated through optimized layout
algorithms, hardware-based graphics acceleration, and a reduced
memory footprint relative to other software. The NAViGaTOR user
interface includes unique tools to help simplify the "hairball",
which is a common challenge in many interaction networks.