Quoting from Moreira, I.S., Fernandes, P.A. & Ramos, M.J. (2010). Protein-protein docking dealing with the unknown. J Comput Chem 31, 317-342:

    "In Figure 3 we have plotted the number of citations per year of the docking programs described earlier (data took from ISI Web of Science considering the references from Table 4). From Figure 3.1 it is possible to observe that only after 2003 there was an increase of the number of citations of the protein-protein docking softwares. Since their publication the most cited softwares are HADDOCK, RosettaDock,three-dimensional-Dock, BIGGER, and Dot. It is possible to observe an increase of the number of citations per year of the Patch-Dock, ClusPro, HADDOCK, RosettaDock and ZDOCK. We have to stress that even though HADDOCK is a recent software, it seems to be very popular presenting a clearly higher number of citations per year. If we consider just papers that apply the different softwares to specific biological problems (represented in Figure 3.2) Haddock is the most popular, followed by ClusPro, PatchDock, and RosettaDock."
    "From the results of the CAPRI experiment and the software popularity we can observe that ICM, ZDOCK, HADDOCK and ROSETTADOCK seem to be some of the best predictors that are most commonly used. Softwares, such as HADDOCK, which are capable of dealing with side-chain and backbone flexibility as well as using biological information regarding the complex in the searching stage, seem to perform better in the protein-protein docking world."