Gene-Level Representation Learning via Interventional Style Transfer in Optical Pooled Screening
This work addresses the need for better gene function analysis in health and disease research, though it appears incremental as it builds on existing style-transfer techniques.
The paper tackled the problem of extracting biologically informative representations from images of genetically perturbed cells in optical pooled screening by using a style-transfer approach to learn gene-level feature representations. The result was that their method outperformed widely used engineered features in clustering gene representations according to gene function.
Optical pooled screening (OPS) combines automated microscopy and genetic perturbations to systematically study gene function in a scalable and cost-effective way. Leveraging the resulting data requires extracting biologically informative representations of cellular perturbation phenotypes from images. We employ a style-transfer approach to learn gene-level feature representations from images of genetically perturbed cells obtained via OPS. Our method outperforms widely used engineered features in clustering gene representations according to gene function, demonstrating its utility for uncovering latent biological relationships. This approach offers a promising alternative to investigate the role of genes in health and disease.