Inferring mechanisms of compensation from genetic interaction networks

Tuesday, February 28, 2012 - 3:00pm - 3:15pm
Keller 3-180
Lenore Cowen (Tufts University)
One of the most well-studied and useful network motifs found
in genetic interaction data is the between-pathway model (BPM),
introduced first by Kelley and Ideker. This is a network motif
consisting of a particular pattern of genetic and physical
interactions that is thought to signify two coherent sets of genes
that may be compensatory or adaptive. We generalize the BPM motif from
unweighted to weighted networks, allowing us to uncover new BPMs based
on the new high-throughput genetic interaction data becoming available
through the use of EMAP and SGA screening methods. We present a
randomized algorithm based on local search for maximal cuts that is
mathematically natural, algorithmically simple, and fast in practice
that can discover these generalized BPMs. We discuss biological
findings and implications of our approach.