Crops such as wheat, barley and rice constitute a major diet of the entire world population. However, they are prone to infection due to pathogens such as fungi, resulting in significant yield losses of crops that could otherwise feed >600 million people. In order to infect crops, plant pathogenic fungi release fungal effector proteins as a form of virulence to mediate infection. Characterising the structure of effector proteins is vital to understanding their virulence mechanisms and the interactions with their hosts. Several effectors have been identified and validated experimentally; however, the lack of sequence conservation among effectors and lack of similarity to known functional proteins often impedes effector identification and prediction of their molecular mechanism of action using sequence similarity approaches. Structural similarity has nonetheless been observed within fungal effector protein families. Since the structure of fungal effector proteins is more conserved than their sequence, there is interest in utilising computational methods to predict their tertiary structure from their sequence. We report the first use of Rosetta ab initio modelling to predict the structure of experimentally-validated effector candidate sequences for which there are no available experimentally-derived structures. The predicted structural models were assessed for structural quality and consensus protein-fold classification, and were found to match known protein folds with known functions. Our findings demonstrate that Rosetta ab initio modelling can be used successfully to predict the structures of effector proteins for which evolutionary function can be inferred.