AI Biotech: Why Your Fastest Path Needs Academic Freedom
AI biotech is moving at light speed. Zera's Bo Wang and Ci Chu explain how to go faster by leaning on academic freedom and open science, not just industry muscle.
🔬Causal Models Need Causal Data - Xaira’s X-Cell model (Bo Wang & Ci Chu)
This episode delves into Zera Therapeutics' AI drug discovery platform, focusing on their X-Cell model. Guests Bo Wang and Ci Chu explain how X-Cell leverages high-throughput causal data generation (Perturb-seq) and a novel diffusion model architecture with biological priors to predict cellular responses to interventions. They discuss the model's generalization capabilities to unseen biological contexts, the current bottlenecks in virtual cell modeling, and the evolving roles of academia and industry in AI for science.
Key takeaways
- Academic freedom isn't a luxury, it's a speed advantage. Bo Wang argues university labs, free from immediate profit pressures, are the wellspring for radically new concepts, like the biological priors behind Zera Therapeutics' X-Cell model. This deep, unconstrained thinking is hard to replicate in industry. Read more →
- Even with advanced AI drug discovery platforms like Zera Therapeutics’ X-Cell model, the current bottleneck for virtual cell modeling isn't just compute power or algorithms, but the type of data we can actually collect. Read more →
- Zera Therapeutics' X-Cell model uses a diffusion language model architecture to predict cellular responses, departing from traditional auto-regressive (GPT-like) methods. Read more →
- Zera Therapeutics' X-Cell AI, a novel diffusion model, predicts how cells respond to interventions using high-throughput causal data (Perturb-seq). Read more →
4 articles from this episode
The Missing Data For True Virtual Cells
Your virtual cell models hit a wall. Bo Wang and Ci Chu pinpoint the radical data breakthroughs needed for true temporal dynamics and functional insights.
X-Cell's Diffusion AI: Refine Gene Prediction, Not Type It
Zera Therapeutics' X-Cell AI reframes gene prediction as iterative "editing," not typing. Learn how diverse biological priors unlock unseen generalization.
Zera's X-Cell AI Predicts Unseen Biology, Beats Baseline
Zera Therapeutics' X-Cell AI just validated predictions in unseen biological contexts, proving true generalization. Founders: This is how AI redefines problem-solving.
More Latent Space episodes
- Inside the Team That Killed Mandatory Code Review — Quinn Slack, AMPSep 13 · 6 articles
- The Inference Frontier: from 100 to 10,000 tokens per second — Sean Lie, Cerebras CTOSep 6 · 6 articles
- Trillion Token Context. No, Really — Anima Anandkumar & Benedikt Jenik, Accelerated UnderstandingSep 6 · 5 articles
- 🔬 Why Transformers Hit a Wall the Moment Physics Shows Up — Anima Anandkumar, CaltechAug 30 · 5 articles