Beta-lactamase proteins are among the most prominent causes of antibiotic resistance. These enzymes confer resistance to beta lactam antibiotics, which are commonly used to treat bacterial infections. In recent years, novel beta-lactamase have emerged exhibiting resistance to all classes of beta lactams, representing a major threat to global health.The mechanism by which beta-lactamase can expand their substrate specificity to confer bacteria with resistance to novel drugs is complex and not-well elucidated. In this work, beta-lactamase function is explored using a variety of computational techniques to identify the molecular mechanisms behind antibiotic resistance. In particular, the connection between protein dynamics and protein function is explored in beta-lactamase, revealing how changes in enzyme motion are related to changes in the enzyme substrate specificity.To study beta-lactamase function, a library of molecular dynamics (MD) simulations was generated which includes simulations of TEM-1, TEM-2, TEM-10, and TEM-52 beta-lactamase, either in its apo or holo form. Holo simulations were performed with the enzymes in complex with either ampicillin, amoxicillin, cefotaxime, or ceftazidime. The enzyme-antibiotic combinations were chosen to represent both wild-type and extended-spectrum beta-lactamase activity.To identify the functional dynamics responsible for substrate recognition, Supervised Projective Learning with Orthogonal Completeness (SPLOC) was employed. SPLOC compared the beta-lactamase simulations under different groupings to understand the role of both enzyme mutations and antibiotic interactions in determining substrate recognition. These motions were also leveraged to classify whether an enzyme would be able to express extended-spectrum antibiotic binding. Finally, the utility of exploiting these functional dynamics to inhibit beta-lactamase function was explored using pepStream. Novel peptides were generated which bound with specificity to regions of the enzyme exhibiting functional dynamics. This work identified dynamic signatures in beta-lactamase underlying substrate recognition. Importantly, these signatures took the form of increased flexibility in loops bordering the active site of the enzymes, which mediate local conformational flexibility that facilitates optimal substrate interactions with different antibiotics. Notably, the dynamic signatures between different protein-antibiotic systems was unique, reflecting the complexity of the mechanisms underlying antibiotic binding. A proof-of-concept for designing de-novo peptides to bind with beta-lactamase at these regions suggests that a novel class of beta-lactamase inhibitors could inhibit these motions required for substrate recognition, yielding a novel method for controlling beta-lactamase mediated antibiotic resistance.