This chapter explores the computational processing of morphological fingerprints for downstream analysis, including compound similarity search and activity prediction. Morphological fingerprints, derived from cell painting assay images, are numerical vectors characterizing the spatial arrangement, morphology, and texture of organelles. Using these vectors, one can train machine learning (ML) models to identify patterns and predict changes in cell morphology upon compound treatment. The chapter consists of three sections, each supported by a Jupyter notebook. The first section covers data preparation for computational analysis, such as ingesting the data, standardization, removing missing values, and normalizing data. The second section details the computation of similarity searches, identifying the closest match to a query, e.g., to identify compounds with a similar mode of action. In addition, structural fingerprints, which are derived from the molecular structure itself, are introduced to perform complementary searches on different molecular fingerprints. The third section demonstrates how to build a basic ML model to predict estrogen receptor activity and provides insights into model tuning, testing, and interpretation. Overall, this chapter provides a comprehensive guide to leveraging morphological fingerprints for advanced computational analysis in drug discovery and activity prediction studies.