If you want to study the existing protein interaction then you can use IntAct, as it include data from most of the databases.

you can also use String which also predict the possible interaction for the proteins, using text mining, homolog information.

There are different other tools which also predict protein interaction based on different features like structure, domain etc. Here is the link for the article, in the table they have summarized the tools. http://www.sciencedirect.com/science/article/pii/S0959440X13001267

You can choose the tool based on the availability of the data.

Good luck!

Another list of available tools based on various types of data (known 3D structures; interaction networks derived from various physical binding and gene expression correlation experiments, etc.) is available at https://www.hsls.pitt.edu/obrc/index.php?page=protein_protein_interactions

It''s not clear from your question what type of organism your protein is derived from. I expect your protein is related to Plant Science. If you are analyzing a protein from a plant pathogen (e.g, from an insect), there should be a wealth of experimentally verified protein-protein interaction (PPI) data available from organisms with similar (physical) PPI networks. If you are working on a plant protein, the currently available experimental data is not as extensive. You may want to look at http://interactome.dfci.harvard.edu/A_thaliana/. PPIs in Arabidopis thaliana are extensively characterized. ( CCSB maintains extensive PPI databases for humans, yeast, viuses, plants and C. elegans.) An extensive analysis of PPIs in maize was recently published. See http://www.plantphysiol.org/content/170/2/618