Proteins perform biochemical functions by forming complexes, or protein-protein interactions (PPIs). Many different approaches such as phage display, yeast hybridization, etc. were developed to illustrate the PPIs, and disclose the composition and organization of protein complexes. However, none of these approaches are based on the real-time and in vivo PPI analysis. Proximity-dependent labeling (PDL) of interacting proteins has recently been proposed by taking advantage of several enzymes, which are capable of attaching the known reactive groups to the nearby proteins covalently. Among the PDL methods, BioID is the earliest and the most widely used one and has been upgraded from its prototype, making it an extremely convenient research tool. In this review, we describe the BioID technology development, its potential applications according to the nature of the target protein, and some recent efforts to circumvent the technical limitations. Moreover, some comparable PDL methods are introduced, including selective proteomic proximity labeling assay using tyramide, enzyme-mediated activation of radical sources, Proximity Labeling with Ascorbate Peroxidase, in vivo proximal labeling, etc., and we propose that systematic comparison of the working radius of these methods may be helpful to develop a tool box, from which the right method can be selected for a given target protein for PPI research.