ProteinPNet: Prototypical Part Networks for Concept Learning in Spatial Proteomics
By: Louis McConnell , Jieran Sun , Theo Maffei and more
Potential Business Impact:
Finds hidden patterns in cancer cells.
Understanding the spatial architecture of the tumor microenvironment (TME) is critical to advance precision oncology. We present ProteinPNet, a novel framework based on prototypical part networks that discovers TME motifs from spatial proteomics data. Unlike traditional post-hoc explanability models, ProteinPNet directly learns discriminative, interpretable, faithful spatial prototypes through supervised training. We validate our approach on synthetic datasets with ground truth motifs, and further test it on a real-world lung cancer spatial proteomics dataset. ProteinPNet consistently identifies biologically meaningful prototypes aligned with different tumor subtypes. Through graphical and morphological analyses, we show that these prototypes capture interpretable features pointing to differences in immune infiltration and tissue modularity. Our results highlight the potential of prototype-based learning to reveal interpretable spatial biomarkers within the TME, with implications for mechanistic discovery in spatial omics.
Similar Papers
Neural Proteomics Fields for Super-resolved Spatial Proteomics Prediction
Quantitative Methods
Makes cell maps show tiny details better.
Topotein: Topological Deep Learning for Protein Representation Learning
Machine Learning (CS)
Helps scientists understand how proteins fold.
Joint Modeling of Spatial Dependencies Across Multiple Subjects in Multiplexed Tissue Imaging
Applications
Finds hidden patterns in cancer cells.