NewLimit combines artificial intelligence with high-throughput biology in a virtuous cycle. Their AI model, Ambrosia, predicts which gene combinations will be effective. These predictions are then tested in thousands of parallel experiments, which in turn generate massive datasets to further train and refine the AI, accelerating discovery.

To break the data bottleneck in AI protein engineering, companies now generate massive synthetic datasets. By creating novel "synthetic epitopes" and measuring their binding, they can produce thousands of validated positive and negative training examples in a single experiment, massively accelerating model development.

Traditional antibody optimization is a slow, iterative process of improving one property at a time, taking 1-3 years. By using high-throughput data to train machine learning models, companies like A-AlphaBio can now simultaneously optimize for multiple characteristics like affinity, stability, and developability in a single three-month process.

A new 'Tech Bio' model inverts traditional biotech by first building a novel, highly structured database designed for AI analysis. Only after this computational foundation is built do they use it to identify therapeutic targets, creating a data-first moat before any lab work begins.

The company focuses on immunology, oncology, and metabolic diseases because their pathways are highly dynamic and require nuanced control. Metaphor’s ability to create antibodies that activate, bias, or multi-target pathways provides a level of precision that simple inhibitors lack.

The key advantage for AI biotech isn't the model itself, but generating massive, proprietary datasets ("science tokens") via automated labs. This novel data, which doesn't exist publicly, is crucial for training superior models and achieving true scientific intelligence.

Building biologically relevant AI is not a one-off process. It demands a continuous "lab in the loop" system where wet lab experiments generate proprietary data to train models, whose outputs are then physically tested in the lab. This iterative feedback cycle constantly refines the model's predictive accuracy.

Traditional drug design crystallizes a receptor to understand its structure, removing it from its biological context. Metaphor reverses this by first studying a receptor's dynamic interactions in living systems, ensuring its antibodies are functionally active from the start.

The platform's generative nature produces a library of viable antibody candidates for a single target, not just one. This optionality is a key advantage, allowing the team to select the molecule with the best combination of potency, developability, and target profile.