Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2501.19178 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913673170124800 |
|---|---|
| author | Kovačević, Luka Gaudelet, Thomas Opzoomer, James Triendl, Hagen Whittaker, John Uhler, Caroline Edwards, Lindsay Taylor-King, Jake P. |
| author_facet | Kovačević, Luka Gaudelet, Thomas Opzoomer, James Triendl, Hagen Whittaker, John Uhler, Caroline Edwards, Lindsay Taylor-King, Jake P. |
| contents | Despite substantial efforts, deep learning has not yet delivered a transformative impact on elucidating regulatory biology, particularly in the realm of predicting gene expression profiles. Here, we argue that genuine "foundation models" of regulatory biology will remain out of reach unless guided by frameworks that integrate mechanistic insight with principled experimental design. We present one such ground-up, semi-mechanistic framework that unifies perturbation-based experimental designs across both in vitro and in vivo CRISPR screens, accounting for differentiating and non-differentiating cellular systems. By revealing previously unrecognised assumptions in published machine learning methods, our approach clarifies links with popular techniques such as variational autoencoders and structural causal models. In practice, this framework suggests a modified loss function that we demonstrate can improve predictive performance, and further suggests an error analysis that informs batching strategies. Ultimately, since cellular regulation emerges from innumerable interactions amongst largely uncharted molecular components, we contend that systems-level understanding cannot be achieved through structural biology alone. Instead, we argue that real progress will require a first-principles perspective on how experiments capture biological phenomena, how data are generated, and how these processes can be reflected in more faithful modelling architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_19178 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | No Foundations without Foundations -- Why semi-mechanistic models are essential for regulatory biology Kovačević, Luka Gaudelet, Thomas Opzoomer, James Triendl, Hagen Whittaker, John Uhler, Caroline Edwards, Lindsay Taylor-King, Jake P. Machine Learning Despite substantial efforts, deep learning has not yet delivered a transformative impact on elucidating regulatory biology, particularly in the realm of predicting gene expression profiles. Here, we argue that genuine "foundation models" of regulatory biology will remain out of reach unless guided by frameworks that integrate mechanistic insight with principled experimental design. We present one such ground-up, semi-mechanistic framework that unifies perturbation-based experimental designs across both in vitro and in vivo CRISPR screens, accounting for differentiating and non-differentiating cellular systems. By revealing previously unrecognised assumptions in published machine learning methods, our approach clarifies links with popular techniques such as variational autoencoders and structural causal models. In practice, this framework suggests a modified loss function that we demonstrate can improve predictive performance, and further suggests an error analysis that informs batching strategies. Ultimately, since cellular regulation emerges from innumerable interactions amongst largely uncharted molecular components, we contend that systems-level understanding cannot be achieved through structural biology alone. Instead, we argue that real progress will require a first-principles perspective on how experiments capture biological phenomena, how data are generated, and how these processes can be reflected in more faithful modelling architectures. |
| title | No Foundations without Foundations -- Why semi-mechanistic models are essential for regulatory biology |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2501.19178 |