Ortholog detection (OD) is a critical first step for a variety of comparative genomic and phylogenetic inference tasks. In general, existing OD methods can be classified as tree-based, graph-based, or a hybrid of the two. Comparative performance between these approaches varies by dataset, however, frustrating attempts to identify a single best method. In contrast, MOSAIC improves ortholog detection by leveraging complementary between diverse OD methods. The result is striking gains in the number of detected orthologs relative to existing approaches, while simultaneously maintaining or improving functional-, phylogenetic-, and sequence identity-based measures of ortholog quality.