CSPOT Tutorial Data

Multiplexed imaging technologies have played a significant role in advancing spatial biology by revealing critical cell communication networks in tissue development and diseases. However, the scalable application of these imaging technologies remains a significant challenge due to the need for human-in-the-loop processes. To address this issue, we developed CSPOT, a computational framework that utilizes a machine learning approach to effectively distinguish real signals from varying background noise and phenotype single cells without visual intervention. CSPOT can learn the features of intercellular protein expression patterns and process highly artifactual images with high accuracy. An end-to-end Python implementation of Gator has been provided to facilitate the large-scale application of spatial biology.