It is of considerable importance to understand molecular mechanisms of human disease and to determine genes responsible. Recently, researchers have begun to use complex cellular networks to address these problems. However, most analyses model proteins as graph-theoretical nodes, ignoring the structural details of individual proteins and the spatial constraints of their interactions. Here, we developed a novel algorithm to investigate on a genomic scale the underlying molecular mechanisms of human genetic disease by integrating 3D atomic-level protein structural genomics information with high-quality large-scale protein interaction data. This network consists of 4,222 high-quality binary protein-protein interactions with their atomic-resolution interfaces. We systematically examine relationships between 3,949 genes, 62,663 mutations and 3,453 associated disorders within the framework of this 3D network. We find that in-frame mutations (missense mutations and in-frame insertions and deletions) are enriched on the interaction interfaces of proteins associated with the corresponding disorders, and that the disease specificity for different mutations of the same gene can be explained by their location within an interface. We also predict 292 candidate genes for 694 unknown disease-to-gene associations with proposed molecular mechanism hypotheses. This work indicates that knowledge of how in-frame disease mutations alter specific interactions is critical to understanding pathogenesis. Structurally resolved interaction networks should be valuable tools for interpreting the wealth of data being generated by large-scale structural genomics and disease association studies.