Simulating large-scale networks of neurons is an important approach both when interpreting and synthesising experimental data from healthy and diseased brains. Here we focus on the striatum, the main input stage and the largest nucleus of the basal ganglia. The basal ganglia are involved in motor learning, action-selection and reinforcement learning. Dysfunction in these areas lead to a variety of brain disorders such as Parkinson’s disease (PD), a neurodegenerative disorder that affects nerve cells. We presented a nearly full-scale data driven model of the mouse striatum using available data on cellular morphology, electrophysiological properties, cell density and synaptic connectivity (Hjorth et al. 2020). To model the progression of Parkinson’s disease we iteratively degenerate the morphologies and estimate the change in connectivity. In particular, neurodegeneration is modelled as a progressive loss (P0, PD1, PD2, PD3) of the most distal fragments of the dendritic arbours of striatal projection neurons (SPN). We also include the compensatory growth of fast spiking (FS) interneuron axons that takes place during early stages of PD degeneration. These processes lead to substantial removal of distal synapses on SPNs, and the addition of some GABAergic FS synapses, mimicking the disease progression. We apply algebraic topology, in particular directed cliques (simplices), to investigate the local structural connectivity in the striatum, and how structural features shape network dynamics. We show that progressive dendritic degeneration not only alters the global connection probabilities but also dramatically affects statistics of simplices, particularly at the last PD stage (PD3). We found that interneurons, despite being the minority, can have a surprisingly large effect on the distribution of simplices. These results suggest that interneurons may play a crucial role in shaping the striatal network structure and dynamics during PD progression.