The aim of the Swedish spotlight program is not only to further our knowledge of the opportunities presented by AI in both healthcare and wider society but also to engage with the challenges and ethical considerations raised by these advances and ensure that our research is grounded in actual experiences and concerns. The Swedish node has selected a number of articles reflecting various aspects of AI in healthcare, including the ethical aspects of using health data needed to train the AI algorithms.
1. Reconstructing the regulatory programs underlying the phenotypic plasticity of neural cancers
Authors: Larsson I, Held F, Popova G, Koc A, Kundu S, Jörnsten R, Nelander S
Institutions Involved: Uppsala University
Summary: This study presents a novel computational method for reconstructing cellular regulatory programs using large-scale single-cell RNA sequencing data from tumours and developing tissues. The algorithm efficiently groups target genes into modules and predicts crucial transcription factors and kinases, all with minimal computational demand. When applied to adult and paediatric brain cancers, the method successfully identified essential regulators and proposed therapeutic interventions, particularly relevant for enhancing drug response in glioblastoma.
2. Artificial intelligence for high content imaging in drug discovery
Authors: Carreras-Puigvert J, Spjuth O
Institutions Involved: SciLifeLab and Uppsala University
Summary: This article explores the transformative role of artificial intelligence (AI) in the field of imaging and microscopy, particularly within high content imaging (HCI). It outlines the shift from traditional image analysis techniques to advanced AI-driven approaches, highlighting their growing impact on drug discovery. AI methods promise cost-effective and accessible means to accelerate the identification of novel therapeutics, with HCI serving as a pivotal experimental platform for these innovations.
3. How to build the virtual cell with artificial intelligence: Priorities and opportunities
Authors: Bunne C, Roohani Y, Rosen Y, Gupta A, Zhang X, Roed M, Alexandrov T, AlQuraishi M, Brennan P, Burkhardt DB, Califano A, Cool J, Dernburg AF, Ewing K, Fox EB, Haury M, Herr AE, Horvitz E, Hsu PD, Jain V, Johnson GR, Kalil T, Kelley DR, Kelley SO, Kreshuk A, Mitchison T, Otte S, Shendure J, Sofroniew NJ, Theis F, Theodoris CV, Upadhyayula S, Valer M, Wang B, Xing E, Yeung-Levy S, Zitnik M, Karaletsos T, Regev A, Lundberg E, Leskovec J, Quake SR
Institutions Involved: SciLifeLab and Royal College of Technology (KTH)
Summary: In this perspective article, the authors propose the concept of building AI Virtual Cells (AIVC) to revolutionise biological sciences through high-fidelity simulations. These virtual models aim to simulate cellular processes in unprecedented detail. Use cases span across disciplines: modelling cancer mutation progression, simulating developmental biology dynamics, and predicting microbiological interactions. This approach promises to reshape scientific understanding and experimentation in biology.
4. Towards an interpretable deep learning model of cancer
Authors: Nilsson A, Meimetis N, Lauffenburger DA
Institutions Involved: SciLifeLab and Karolinska Institutet
Summary: The authors introduce a deep learning framework that models complex molecular interactions underlying cancer development. By embedding prior biological knowledge into a recurrent neural network architecture, the model can capture dynamic cellular behaviours. Training on omics data across diverse perturbations allows for insightful representations of gene regulation, signalling, and metabolic pathways, potentially leading to more interpretable and predictive cancer models.
5. Ethical and social reflections on the proposed European Health Data Space
Authors: Staunton C, Shabani M, Mascalzoni D, Mežinska S, Slokenberga S
Institutions Involved: Uppsala University
Summary: This article critiques the proposed European Health Data Space (EHDS), a regulatory initiative introduced to address the fragmented legal landscape surrounding health data sharing in the EU. While acknowledging the need for reform, the authors stress that legal changes must align with ethical principles and the rights of data subjects. The discussion includes an analysis of GDPR-related complexities and offers thoughtful recommendations for improving data governance frameworks in health research.
