Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models

TL;DR

This paper demonstrates that combining sparse dictionary learning with PCA whitening enables the extraction of biologically meaningful concepts from microscopy foundation models, advancing scientific discovery in bioimaging.

Contribution

It introduces a novel method, ICFL with PCA whitening, to extract interpretable biological concepts from microscopy models, a task previously limited by data complexity.

Findings

Successfully retrieved cell types and genetic perturbations

Revealed subtle morphological changes from interventions

Enhanced interpretability of microscopy foundation models

Abstract

Sparse dictionary learning (DL) has emerged as a powerful approach to extract semantically meaningful concepts from the internals of large language models (LLMs) trained mainly in the text domain. In this work, we explore whether DL can extract meaningful concepts from less human-interpretable scientific data, such as vision foundation models trained on cell microscopy images, where limited prior knowledge exists about which high-level concepts should arise. We propose a novel combination of a sparse DL algorithm, Iterative Codebook Feature Learning (ICFL), with a PCA whitening pre-processing step derived from control data. Using this combined approach, we successfully retrieve biologically meaningful concepts, such as cell types and genetic perturbations. Moreover, we demonstrate how our method reveals subtle morphological changes arising from human-interpretable interventions, offering a promising new direction for scientific discovery via mechanistic interpretability in bioimaging.