Given the growing scarcity of sustainable bioresources, the biggest challenge is not only finding sufficient volume but also managing the complex variability. The international consortium Project OPAL presented groundbreaking results in which AI models and robotics take control in the laboratory. This maximizes the use of scarce bioresources and accelerates the large-scale production of green chemicals and fuels.
In the petrochemical industry, stability is the norm: a refinery is set to deliver a constant flow of crude oil with a known composition. However, bio-based feedstocks, ranging from agricultural waste streams to woody biomass, are volatile. Their chemical composition varies with harvest, region, and season. Until now, this meant that chemical reactors required constant manual adjustment, leading to significant yield losses and high costs.
The Genesis Mission: AI That Understands Biology
The OPAL project (Orchestrated Platform for Autonomous Laboratories) is part of a prestigious new initiative by the US Department of Energy: the Genesis missionAt the heart of this is the ModCon consortium, which focuses on building foundational AI models for biology.
While well-known AI systems like ChatGPT are trained on text, OPAL's models are trained on enormous, complex datasets of genomes, proteins, and metabolic functions. According to project leader Paramvir Dehal of Berkeley Lab, this was previously the biggest bottleneck: biological data is often messy and organized differently. OPAL now uses supercomputers to create the largest and most precise biological datasets ever, allowing the AI to learn to understand the "language" of microorganisms.
Autonomous laboratories in practice
OPAL's system operates as a closed loop, minimizing human intervention:
- Direct analysis: The AI analyses the chemical fingerprint of an incoming batch of residual flows.
- Prediction and design: The model predicts which microorganism is most suitable for this specific batch and, if necessary, designs genetic adjustments to increase yield.
- Robotic execution: Autonomous systems in the laboratory physically perform the experiments and feed the results directly back to the AI to refine the process.
From fighter jet fuel to critical minerals
The impact of this technology extends beyond basic chemistry. Paul Adams, Associate Laboratory Director at Berkeley Lab, emphasizes that this AI-driven approach is crucial for the production of high-quality aviation biofuels (jet fuel precursors) and even for the biorecovery of critical minerals.
By reducing the time between research and commercial application by an estimated 40%, OPAL offers a solution to the scarcity of bio-based feedstocks. It enables the industry to convert previously unsuitable "low-value" streams into valuable products with maximum efficiency.
Source: Berkeley Lab News Center: “Foundational AI Models to Accelerate Biological Discovery”
Photo: sorapop, Adobe Stock
