Parkinson’s disease (PD) is a progressive neurodegenerative disorder of the central nervous system, characterized by motor symptoms that are primarily associated with the loss of dopaminergic neurons (DA). These cells are primarily found in the substantia nigra, which is responsible for the initiation and modulation of involuntary movements, cognitive planning, and other vital functions. Single-nucleus RNA sequencing (snRNA-seq) is a widely used technique for profiling the cellular composition of brain tissues, allowing the study of alterations in cell composition under various conditions. However, the accessibility of human brain samples remains a challenge. In this study, we analyzed changes in the composition of ten DA subtypes in both control and Parkinson’s disease (PD) brain samples using CIBERSORTx, a machine-learning deconvolution method. The analysis integrated single-nucleus RNA sequencing (snRNA-seq) reference data with bulk RNA sequencing (RNA-seq) data to quantify subtype-specific alterations. The results showed the composition of six out of ten DA subtypes on control and PD samples, highlighting the presence of SOX6_DDT with the highest abundance. The methodological approach used in this work suggests a method for analyzing shifts in cell subtype (or types) composition for diseases caused by cell loss (such as Parkinson’s disease) without the need for physical isolation of the cells.