MORPHIS (MORPHological Interpretable Signature) captures heterogeneous treatment- and aging-related responses of single cells
Bohr, F.; Oikonomou, A.; Nielsen, E. E. M.; Konstantinidis, G.; Mortensen, J. S.; Tavernarakis, N.; Nielsen, H. M.; Hatzakis, N. S.
Show abstract
Cell morphology encodes changes in cytoskeletal and organelle organization during disease, treatment, and aging, yet is often assessed qualitatively or through poorly interpretable feature sets extracted from microscopy images. Here we introduce MORPHIS (MORPHological Interpretable Signature), a machine learning framework on explainable, analytically rich features paired with statistical methodologies for robust and interpretable quantification of single-cell morphological signature. MORPHIS extracts compact, interpretable feature signatures that capture both perturbation-specific response magnitude and heterogeneous cellular responses. It accurately distinguishes treatment-specific morphological signatures of eight mechanistically distinct membrane-active and intracellular-targeting compounds in Caco-2 and HeLa cells elucidating conserved and divergent phenotypic responses among compound-classes, as well as ultrastructural nuclear alterations upon aging of C. elegans and quantifies heterogeneous single-cell fractional responses. By remaining cell-type and perturbation agnostic, MORPHIS provides a generalizable framework for quantifying morphological signatures across diverse biological contexts including pharmacological treatment or aging.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Transformer-based spatial-temporal detection of apoptotic cell death in live-cell imaging. 95%
- An Image-Based Data-Driven Analysis of Cellular Architecture in a Developing Tissue 94%
- MiSiC, a general deep learning-based method for the high-throughput cell segmentation of complex bacterial communities 93%
Similar papers in this journal
Similar papers in this journal
- Supervised and unsupervised deep learning-based approaches for studying DNA replication spatiotemporal dynamics 95%
- Multiparametric quantitative phase imaging for real-time, single cell, drug screening in breast cancer 93%
- Large Scale Cell Painting Guided Compound Selection Reveals Activity Cliffs and Functional Relationships 93%
Similar papers in this journal
- KDML: a machine-learning framework for inference of multi-scale gene functions from genetic perturbation screens 95%
- Machine learning inference of continuous single-cell state transitions during myoblast differentiation and fusion 93%
- PIFiA: Self-supervised Approach for Protein Functional Annotation from Single-Cell Imaging Data 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.