Latent Space discusses Fabian Theis on AI for Virtual Cell modeling
Original title: Controlling Biology - Fabian Theis, Helmholtz Munich
Latent Space interviews Fabian Theis about $1.8B Biohub funding for multi-scale biological data and challenges in perturbation prediction.
Latent Space covers Fabian Theis' leadership in Virtual Cell AI research at Helmholtz Munich. The piece reviews recent $1.8B Biohub commitments from Biohub, NIH, DOE, Google and Meta for new imaging and data capabilities. Theis notes that advanced multi-scale data could address scale interfaces in biology but highlights that current models often fail basic perturbation prediction tasks.
Unconfirmed: The $1.8B figure includes prior NIH investments
- Biohub announced $1.8B in commitments for biological data collection
- Perturbation prediction models frequently fail to beat simple baselines
- Improved metrics allow deep learning models to outperform mean predictions
- Focus on multi-scale predictions from molecules to organisms
- Helmholtz integrates AI with precision medicine data
From the original article
4 min read at latent.space
Fabian Theis has been one of the key players in the various iterations of the Virtual Cell concept and initiatives. It is not hyperbole at all to say that his lab has been pushing the frontier of algorithms to understand multi-scale, multi-modal (”mult...
By Latent.Space · latent.space
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.