Interpretability analysis on a pathology foundation model reveals biologically relevant embeddings across modalities

Mechanistic interpretability has been explored in detail for large language models (LLMs). For the first time, we provide a preliminary investigation with similar interpretability methods for medical imaging. Specifically, we analyze the features from a ViT-Small encoder obtained from a pathology Foundation Model via application to two datasets: one dataset of pathology images, and one dataset of pathology images paired with spatial transcriptomics. We discover an interpretable representation of cell and tissue morphology, along with gene expression within the model embedding space. Our work paves the way for further exploration around interpretable feature dimensions and their utility for medical and clinical applications.

Inference of cell dynamics on perturbation data using adjoint sensitivity

Data-driven dynamic models of cell biology can be used to predict cell response to unseen perturbations. Recent work (CellBox) had demonstrated the derivation of interpretable models with explicit interaction terms, in which the parameters were optimized using machine learning techniques. While the previous work was tested only in a single biological setting, this work aims to extend the range of applicability of this model inference approach to a diversity of biological systems. Here we adapted CellBox in Julia differential programming and augmented the method with adjoint algorithms, which has recently been used in the context of neural ODEs. We trained the models using simulated data from both abstract and biology-inspired networks, which afford the ability to evaluate the recovery of the ground truth network structure. The resulting accuracy of prediction by these models is high both in terms of low error against data and excellent agreement with the network structure used for the simulated training data. While there is no analogous ground truth for real life biological systems, this work demonstrates the ability to construct and parameterize a considerable diversity of network models with high predictive ability. The expectation is that this kind of procedure can be used on real perturbation-response data to derive models applicable to diverse biological systems.