Artificial Intelligence-based Morphological Profiling for Drug Discovery: Advancing Towards 4D Phenotypic Profiling

Margreet R. de Vries, Lucas Dent, Hugh Sparks, Reed Naidoo, Olga Fourkioti, Mar Arias-García, Christopher Dunsby, Chris Bakal, Matt De VriesOpenalex Unknown 2025 · DOI · OpenAlex

Abstract

Phenotypic profiling methods for drug discovery have received revitalised interest due to the rapid adoption of novel computational methods, including artificial intelligence-based techniques. However, these methods predominantly analyse 2D images of 2D cell cultures, which can result in suboptimal predictive validity. This chapter highlights the transition to 4D morphological profiling, which integrates time-resolved 3D imaging of 3D cell cultures to capture cellular morphodynamics. We explore the evolution of morphological profiling and emphasise the pivotal role of deep learning in enhancing the resolution and depth of cellular analysis. Through detailed discussions on the historical background, current applications, and future directions of morphological profiling, we suggest how 4D phenotypic profiling offers a more accurate representation of cellular dynamics, thus potentially revolutionising drug discovery and therapeutic development.
BibTeX
@inproceedings{vries2025artificial,
  author = {Margreet R. de Vries and Lucas Dent and Hugh Sparks and Reed Naidoo and Olga Fourkioti and Mar Arias-García and Christopher Dunsby and Chris Bakal and Matt De Vries},
  title = {{Artificial Intelligence-based Morphological Profiling for Drug Discovery: Advancing Towards 4D Phenotypic Profiling}},
  booktitle = {Openalex Unknown},
  year = {2025},
  doi = {10.1039/9781837676941-00209},
}

Loading related content...

Figure