By applying complex network analysis, graph algorithms, and deep learning, we aim to develop approaches to better interpret the massive and ever-growing protein sequence, structure, and function data in public repositories.