Single-cell mass spectrometry-based proteomics (SCP) provides researchers with high-throughput protein quantification at the single-cell level, which is revolutionizing the view on complex biological processes and disease. SCP has a pivotal advantage over single-cell gene expression technologies by directly assessing proteins, key switches that play vital roles in cell proliferation, migration, metastasis and ageing. However, extracting biological knowledge from SCP data is hampered by the current state-of-the-art data analysis practices that do not correctly account for 1) strong batch effects and missing values, 2) correlation in the abundance of proteomic features between cells of the same sample, and 3) only infer differences in average abundance, limiting the benefits of SCP compared to bulk approaches. In this ambitious project, we will therefore develop cutting-edge data analysis solutions for SCP that 1) provide simultaneous dimension reduction and imputation of missing values while correcting for batch effects, 2) correctly address the hierarchical correlation structure of SCP data to infer reliable proteomic signatures, 3) not only infer on shifts in mean but also in detection rates, variance, skewness and multimodality. These developments are essential to infer the effects of long-term drug treatment and drug resistance from SCP studies of the Prof. Gevaert lab, which we will use to illustrate the relevance and impact of our tools.