Ribosomally-synthesized and post-translationally modified peptides (RiPPs) form an important class of structurally and functionally diverse natural products with tremendous potential as potent and promising drug leads. RiPPs are composed of a leader peptide and core peptide; modifying enzymes recognize the leader and modify residues in the core peptide to generate the active product. These peptides are classed based on their post-translational modification, which confer a large degree of natural diversity and potent bioactivities, with 31 defined families. One family of RiPPs are ‘graspetides,’ which are modified by ATP-grasp enzymes; these modifications typically involve the formation of unique macrolactone and macrolactam ring structures which can confer a high degree of stability and target specificity., seen in microviridins and omega-ester peptide families. For example, the compound Microviridin J is cyclized by ATP-grasp enzymes to form a tricyclized product with remarkable stability and low nanomolar inhibitory activity against serine proteases. These ATP-grasp enzymes have essential functions in primary and secondary metabolism and are defined by a unique structural fold which binds ATP. Despite the strongly structurally conserved ATP-grasp fold, there is a large degree of variety in primary sequences of the enzymes, as well as across the modified substrates. The current understanding of ATP grasp enzymes remains compartmentalized and limited within the field, as the central topic review for ATP-grasp enzymes has not been updated in 9 years. In this time, not only have new subclasses of ATP-grasp enzymes been realized, but new scientific technology, particularly developments in computational chemistry and predictive modeling, have increased the demand for accurate consensus structures. Many current genome mining techniques and programs rely on outdated or incomplete consensus information to predict the location of new RiPPs peptides based on proximity to putative modifying enzymes. This renders inaccurate results or produces peptides vastly similar to what is already known. It is imperative to redefine the ATP-grasp enzyme in light of new technologies and discoveries. In this project, we aim to improve understanding of ATP-grasp enzymes in both primary and secondary metabolism systems. By reevaluating and generating not only a consensus sequence but structure for ATP-grasp enzymes we will be able to more accurately locate and identify RiPPs in the genome mining process. Our central hypothesis is there is a direct correlation between the sequence variations and the substrate variations; by generating a consensus structure using enzymes with a variety of substrates and developing an enhanced understanding of the protein folding process, as well as elucidating the structure function relationship, a predictive ‘template’ for ATP grasp enzymes can be developed. Future directions of this project include the application of the consensus structure to a deep learning pipeline utilizing hipergator and NIVIDIA software to identify RiPPs. The consensus will be used to locate putative ATP-grasp enzymes then proximity and BCG considerations will predict probable graspetides. Once these new graspetides are located they can then be screened for activity and medicinal usage.