PENGUIN: A rapid and efficient image preprocessing tool for multiplexed spatial proteomics

Abstract

Multiplex spatial proteomic methodologies can provide a unique perspective on the molecular and cellular composition of complex biological systems. Several challenges are associated to the analysis of imaging data, in particular regarding the normalization of signal-to-noise ratios across images and background noise subtraction. However, straightforward and user-friendly solutions for denoising multiplex imaging data that are applicable to large datasets are still lacking. We have developed PENGUIN -Percentile Normalization GUI Image deNoising: a rapid and efficient image preprocessing tool for multiplexed spatial proteomics. In comparison to existing approaches, PENGUIN stands out by eliminating the need for manual annotation or machine learning model training. It effectively preserves signal intensity differences and reduces noise, thereby enhancing downstream tasks like cell segmentation and phenotyping. PENGUINs simplicity, speed, and user-friendly interface, deployed both as script and as a Jupyter notebook, facilitate parameter testing and image processing. We illustrate the effectiveness of PENGUIN by comparing it with conventional image processing techniques and solutions tailored for multiplex imaging data. This comparison underscores PENGUINs capability to produce high-quality imaging data efficiently and consistently.