By combining high-content imaging with gene expression profiling at single cell resolution, the team led by Associate Professor Helena Kilpinen uncovered how the morphology of brain cells reflects dynamic cellular processes and even heritability of traits like schizophrenia and bipolar disorder.
Cell Painting method uses fluorescent dyes to visualize key organelles like mitochondria, the endoplasmic reticulum and the nucleus. In the study, published in
They profiled over 60,000 stem cell-derived cortical neurons across three developmental timepoints. By modeling the relationship between image-based features and gene expression, they created a proof-of-concept framework that links what cells look like with biological processes underlying diseases.
“This is the first time, to our knowledge, that cell morphology phenotypes have been used to interpret the possible function of disease-associated genetic variants,” said first author Adithi Sundaresh, a Doctoral Researcher at the Institute for Molecular Medicine Finland (FIMM).
“We believe this could be a powerful approach to bridging the gap between genome-wide association study findings and the biological processes underlying complex diseases.”
Morphology meets molecular biology
Importantly, Cell Painting was able to distinguish broad neuronal classes and track dynamic biological processes like mitochondrial function and neuronal differentiation. These are features that are often missed by gene expression data alone.
“Today’s technologies can produce vast quantities of single-cell data. By combining two of them, imaging and gene expression profiling, we demonstrate how these approaches can make each other easier to interpret,” said the shared first author Dimitri Meistermann from the University of Helsinki.
The study also showed that genes associated with cell morphological features could capture heritability for brain-related traits such as schizophrenia, ADHD and bipolar disorder. This opens up new possibilities for using image-based cellular features to interpret genetic risk in complex diseases.
To test the disease relevance of their model, the team applied their joint profiling approach to patient-derived cells from individuals with Kabuki Syndrome, a rare neurodevelopmental disorder. The data revealed subtle but significant shifts in cell state, highlighting the potential of this method for identifying disease-specific cellular phenotypes.
The authors suggest that this multi-modal approach could become a valuable tool in disease modelling and precision medicine.
“We have been looking for a scalable assay that can help us map how disease variants and genes impact broader cellular functions, beyond the transcriptome. Cell Painting is very well-suited for this purpose, especially in cells with complex morphology, such as neurons,” said Helena Kilpinen.
The research was conducted at the Helsinki Institute of Life Science (HiLIFE) and the Institute for Molecular Medicine Finland (FIMM) and Neuroscience Center (NC), with contributions from the Faculty of Medicine and the Faculty of Biological and Environmental Sciences at the University of Helsinki.
Original publication: Joint profiling of cell morphology and gene expression during in vitro neurodevelopment. Adithi Sundaresh, Dimitri Meistermann, Riina Lampela, Zhiyu Yang, Rosa Woldegebriel, Andrea Ganna, Pau Puigdevall, Helena Kilpinen. eLIFE 1 Dec 2025.