The broad goal of this thesis work is to reveal the structural elements of peripheral membrane-binding proteins, specifically the Transmembrane immunoglobulin and mucin domain (TIM) protein family, that directly confer sensitivity to phospholipid membranes. Peripheral membrane-binding proteins can distinguish their target membranes from those of different cells or even of the same cells in different states. This specific recognition is only possible due to the varying membrane compositions associated with each organelle, type of cell, or state of a cell. Therefore, peripheral membrane-binding proteins must be structurally “fine-tuned” to the physical and chemical properties of their target membranes. While the structures of many phospholipid-binding proteins have been determined in solution, the membrane-bound structures of these proteins are typically intuited assuming that the membrane induces no conformational change in the protein. Alternatively, these structures are simulated with no direct experimental constraints and are indirectly validated in their comparison with experimental results. Unlike structures of protein-protein interactions, which isolate specific molecular interactions, protein-membrane models lack such molecular details. Without this information, it is exceedingly difficult to determine how disease variants of these proteins disrupt membrane-binding or to design proteins with a desired membrane specificity. The field demands a rigorous method for determining the membrane-bound states of peripheral membrane-binding proteins and thus the protein residues which directly interact with the membrane. The largest contributions of this thesis are the development of such a method and its application to the recognition of phosphatidylserine (PS) by three murine members of the TIM protein family, TIM1, TIM3, and TIM4. In this dissertation, the structures of the TIM proteins are used to correlate differences in their association with PS-containing membranes to the residues that interface with the lipid membrane. These models implicate specific energetic contributions to protein-membrane binding, such as hydrophobic insertion and electrostatic engagement. The relative strength of each of these energetic terms correlates with the selectivity of the TIM proteins for particular membranes. Moreover, the membrane-bound structures identify the amino acid residues that directly implement these interactions. This work provides both the molecular details that underlie the recognition of PS-containing membranes by the TIM proteins and the prescription for obtaining such molecular details for the broader class of peripheral membrane-binding proteins.