Interpretable Representation Learning for 3D Multi-Piece Intracellular Structures Using Point Clouds

A key challenge in understanding subcellular organization is quantifying interpretable measurements of intracellular structures with complex multi-piece morphologies in an objective, robust and generalizable manner. Here we introduce a morphology-appropriate representation learning framework that uses three-dimensional rotation-invariant autoencoders and point clouds. This framework is used to learn representations of complex shapes that are independent of orientation, compact and interpretable. We apply our framework to intracellular structures with punctate morphologies (for example, DNA replication foci) and polymorphic morphologies (for example, nucleoli). We explore the trade-offs in the performance of this framework compared to image-based autoencoders by performing multi-metric benchmarking across efficiency, generative capability and representation expressivity metrics. We find that the proposed framework, which embraces the underlying morphology of multi-piece structures, can facilitate the unsupervised discovery of subclusters for each structure. We show how this approach can also be applied to phenotypic profiling using a dataset of nucleolar images following drug perturbations.