Inferring the composition of cell types within complex biological samples can reveal valuable insights into underlying biological processes. While transcriptomics data has been extensively utilized for such analyses, the application of proteomics data remains relatively underexplored. This project aims to evaluate computational approaches for estimating cell-type composition using proteomics data from bulk samples. By leveraging experimentally defined cell populations and simulated mixtures, we will systematically assess the effects of preprocessing methods and computational tools on deconvolution outcomes. Preliminary results indicate the potential of proteomics for accurate cell-type deconvolution under optimized conditions. This project will also develop and provide computational tools to streamline proteomics data preprocessing and analysis, paving the way for broader adoption of proteomics in cell composition studies. Access to high performance computing resources is critical to execute large-scale simulations and analyses, which will enable a deeper understanding of the methodologies and enhance the reproducibility of the findings.