 CREOelectron microscopy, CREOEM, is a powerful technique for determining the structures of biological macromolecular complexes. However, the widely used template-based particle-picking process is labor-intensive and time-consuming. To address this bottleneck, researchers have developed CREOPP, a large, diverse, expected CREOEM image dataset for protein particle-picking and analysis. This dataset contains 9,893 high-resolution micrographs with labeled protein particle coordinates, which will help accelerate the development of both artificial intelligence, AI, and classical methods for automated CREOEM protein particle-picking. This article was authored by Ashwinder Cole, Rajan Jayawali, Ligwa Wang, and others.