The company's core technology, AlphaSeq, uses engineered yeast mating as a proxy for protein binding. The rate of mating corresponds to the binding affinity of proteins on the cell surfaces. By sequencing the resulting cells, the company can count genetic barcodes to quantitatively measure millions of protein-protein interactions at once.
Related Insights
The power of AI for Novonesis isn't the algorithm itself, but its application to a massive, well-structured proprietary dataset. Their organized library of 100,000 strains allows AI to rapidly predict protein shapes and accelerate R&D in ways competitors cannot match.
The cost to generate the volume of protein affinity data from a single multi-week A-AlphaBio experiment using standard methods like surface plasmon resonance (SPR) would be an economically unfeasible $100-$500 million. This staggering cost difference illustrates the fundamental barrier that new high-throughput platforms are designed to overcome.
Gordian Biotechnology embeds unique genetic "barcodes" into hundreds of different gene therapies. This transforms gene therapy from a treatment modality into a high-throughput screening tool, allowing them to test many potential drugs simultaneously inside a single living animal and trace which ones worked.
Contrary to the popular belief that antibody development is a bespoke craft, modern methods enable a reproducible, systematic engineering process. This allows for predictable creation of antibodies with specific properties, such as matching affinity for human and animal targets, a feat once considered a "flight of fancy."
The next leap in biotech moves beyond applying AI to existing data. CZI pioneers a model where 'frontier biology' and 'frontier AI' are developed in tandem. Experiments are now designed specifically to generate novel data that will ground and improve future AI models, creating a virtuous feedback loop.
To overcome on-target, off-tumor toxicity, LabGenius designs antibodies that act like biological computers. These molecules "sample" the density of target receptors on a cell's surface and are engineered to activate and kill only when a specific threshold is met, distinguishing high-expression cancer cells from low-expression healthy cells.
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.
AlphaFold's success in identifying a key protein for human fertilization (out of 2,000 possibilities) showcases AI's power. It acts as a hypothesis generator, dramatically reducing the search space for expensive and time-consuming real-world experiments.
Profluent CEO Ali Madani frames the history of medicine (like penicillin) as one of random discovery—finding useful molecules in nature. His company uses AI language models to move beyond this "caveman-like" approach. By designing novel proteins from scratch, they are shifting the paradigm from finding a needle in a haystack to engineering the exact needle required.
Antibodies bind to specific amino acid sequences, making them unable to distinguish between a protein's healthy and toxic structural forms. Alt-Pep's synthetic peptides use a complementary structure (alpha-sheet) to selectively bind only the toxic oligomers, enabling both targeted therapy and highly specific diagnostics.
