Formulation of drugs as cocrystals  offers an opportunity to modulate  physical  properties, so identification of cocrystal formers for an active pharmaceutical ingredient is an area of great relevance  to the pharmaceutical industry. Exhaustive experimental cocrystal screening is a  very time-consuming task but we have developed a computational method for identifying cocrystal formers that have a high probability of success based on calculated functional group interaction energies. The advantage of our approach is that it is sufficiently  fast to be used as a  high  throughput  method to screen very large numbers of potential coformers. The virtual  screening tool has been applied to several APIs with a large coformer library in order to experimentally validate the method. Our results suggest that success rates in cocrystal  screening  can be significantly  improved by application of computational filters to select the most appropriate coformers for experimental study