| Abstract: | Synthetic genetic arrays have been very effective at measuring genetic interactionsin yeast in a high-throughput manner and recently have been expanded to measurequantitative changes in interaction, termed ''differential interactions'', acrossmultiple conditions. Here, we present a strategy that leverages statisticalinformation from the experimental design to produce a novel, quantitativedifferential interaction score, which performs favorably compared to previousdifferential scores. We also discuss the added utility of differentialgenetic-similarity in differential network analysis. Our approach is preferred fordifferential network analysis, and our implementation, written in MATLAB, can befound at http://chianti.ucsd.edu/~gbean/compute_differential_scores.m. |
