Parkinson’s disease (PD) is a debilitating neurological disorder that affects more than 6.2 million people globally. Although there is no proven stage-modifying treatment, several methods, including Deep Brain Stimulation (DBS), have been developed to control the symptoms of PD by delivering electrical stimulations into the subcortical regions. Despite successful outcomes of DBS, its underlying mechanism remained unknown. Recent experimental studies showed that DBS can induce short-term synaptic plasticity (STP) at the stimulation targets. However, to fully decipher the role of STP in the underlying mechanism of DBS, it is required to infer the dynamics of DBS-induced STP. In this thesis, I developed a new computational framework to study and an inference algorithm for the synaptic dynamics during DBS from intra- and extracellular recordings of the brain activities during DBS.