Measurement of mechanical forces of cell-cell interactions is important for studying the emergence of diverse three-dimensional morphologies of multicellular organisms. We previously reported an image-based statistical method for inferring effective pairwise forces of cell-cell interactions (i.e., attractive/repulsive forces), where a cell particle model was fitted to cell tracking data acquired by live imaging. However, because the particle model is a coarse-grained model, it remains unclear how the pairwise forces relates to sub-cellular mechanical components including cell-cell adhesive forces. Here we applied our inference method to cell tracking data generated by vertex models that assumed sub-cellular components. Through this approach, we investigated the relationship between the effective pairwise forces and various sub-cellular components: cell-cell adhesion forces, cell surface tensions, cell-extracellular matrix (ECM) adhesion, traction forces between cells and ECM, cell growth, etc. We found that the cell-cell adhesion forces were attractive, and both the cell surface tensions and cell-ECM adhesive forces were repulsive, etc. These results indicate that sub-cellular mechanical components can contribute to the effective attractive/repulsive forces of cell-cell interactions. This comprehensive analysis provides theoretical bases for linking the pairwise forces to the sub-cellular mechanical components: this showcase is useful for speculating the sub-cellular mechanical components from the information of cell positions, and for interpreting simulation results based on particle models.