The advent of increasingly sophisticated imaging platforms has allowed for the visualization of the murine nervous system at single-cell resolution. However, current cell counting methods compromise on either the ability to map a large number of different cell types, or spatial coverage, and therefore whole-brain quantification of finely resolved neural cell subtypes remains elusive. Here we present a comprehensive and novel computational pipeline called Matrix Inversion and Subset Selection (MISS) that aims to fill this gap in knowledge and has the ability to infer counts of diverse collections of neural cell types at sub-millimeter resolution using a combination of single-cell RNAseq and in situ hybridization datasets. We rigorously demonstrate the accuracy of MISS against literature expectations, and in so doing, provide the first verified maps for twenty-five distinct neuronal and nonneuronal cell types across the whole mouse brain. The resulting comprehensive cell type maps are uniquely suited to address important open questions in neuroscience. As a demonstration, we utilize our inferred maps to quantitatively establish that adult cell type distributions reflect the ontological splits between brain regions during neural development, indicating that the developmental history of brain regions informs their final cell-type composition. Although this link has been frequently surmised by neuroscientists, its quantitative verification across the entire brain has not previously been demonstrated. Together, our results suggest that the MISS pipeline can be used to generate accurate spatial quantification of diverse cell types without the direct imaging of known type-specific markers, as has been done previously. The entire MISS pipeline and outputs will be made open source in order to catalyze future neuroscientific advances across disciplines.