We scan new podcasts and send you the top 5 insights daily.
Younger notes the PDB shows how proteins physically interact but not how strongly, and it has almost no examples of antibodies one or two mutations apart with measured affinities. Zero-shot optimization instead needs local affinity landscapes built from many single-point mutation variants across many targets.
To evolve AI from pattern matching to understanding physics for protein engineering, structural data is insufficient. Models need physical parameters like Gibbs free energy (delta-G), obtainable from affinity measurements, to become truly predictive and transformative for therapeutic development.
While de-duplicating protein databases helps learn diverse structures, subtle variations in similar sequences are essential for learning function, as a single mutation can be catastrophic. This justifies training models on massive, unclustered datasets to capture fine-grained functional determinants.
Machine learning's application in multi-specific antibody design is hampered by a lack of public data. Companies must invest heavily in generating their own large-scale, proprietary datasets to train effective models, creating a significant barrier to entry and a competitive advantage.
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.
Younger says model makers evaluate on unique bespoke targets, so pharma buyers cannot tell which model is best. He is working with companies on a benchmark that standardizes target difficulty and characterizes tens of thousands of designs, hoping to get true apples-to-apples comparisons with tighter error bars.
Current AI for protein engineering relies on small public datasets like the PDB (~10,000 structures), causing models to "hallucinate" or default to known examples. This data bottleneck, orders of magnitude smaller than data used for LLMs, hinders the development of novel therapeutics.
Younger says de novo designs currently achieve reasonable single-digit percent success rates only against the easiest targets and epitopes. The field's ambition is to engineer challenging epitopes and additional properties, where success rates remain very low; he argues more and better data are still needed.
Standard co-folding models struggle with induced proximity because public databases like the PDB lack sufficient training data for multi-protein complexes. Because dynamic multi-protein complexes take on endless conformations, standard AI models become overconfident and inaccurate. To solve this, Protai generates proprietary structural proteomics data that maps physical distances and conformations, enabling accurate modeling of complexes that can be targeted by molecular glues and PROTACs.
David Younger says his company's model training showed that solving truly hard protein design problems will require much more data than the field expects. He also warns public binding and affinity data are often not interoperable, lack proper controls, and lack annotations, making them hard to reuse.
Younger says the Protein Data Bank today holds about 10,000 antibody-antigen structures and adds roughly 1,000 yearly. Their internal scaling work suggests good generalizable de novo design models may need hundreds of thousands to millions of structures, so waiting on PDB growth is far too slow.