Fatty acids are important sources of energy and possible predictors and etiologic factors in many common complex pathologies such as cardiovascular disease, diabetes, and certain forms of cancers. While fatty acids are thought to covary with each other, their underlying causal networks have not been fully elucidated. This study reports the identification and analysis of a statistical causal network among 15 mostly long-chain fatty acids. In an African-American population sample and using the Genome granularity-Directed Acyclic Graph (GDAG) algorithm, we determined directions or causal relationships in the fatty acid metabolome. A directed causal network was constructed that revealed 29 significant edges among the 15 nodes (p < 0.001). We report that two fatty acid metabolites, palmitoleate and margarate, which originate from dietary intake, together influence every other fatty acid in the network. On the other hand, despite its high connectivity, dihomo-linoleate did not appear to play an important role over the whole fatty acid network. These findings collectively suggest possible strategic entry points for new treatments or preventive modalities against diseases affected by fatty acid metabolites such as cardiovascular disease, diabetes, and obesity. Further studies examining the embedded substructure of the fatty acid metabolite networks in independent population samples would be timely and warranted as we move toward novel postgenomic diagnostics and therapeutics.