AI/ML-assisted antibody engineering - revolution or hype?
01 Dec 2025
AI & Machine Learning
- How does, or will, AI / ML improve antibody engineering? Will it become faster or more precise? How can we metricise that improvement?
- AI assisted antibody engineering will only be as good as the training data sets used to train the models. How do we ensure that the data sets are accurate? What data and metadata sets should we use? (sequences, structures, affinity, clinical responses)
- Can AI/ML models ever replace wet lab engineering? What quality control does occur on the in silico predictions, can it be done without wet lab tests? How can we ensure the models are focusing on the correct biological issue? Do ML engineers and biologists speak the same language enough to ensure that?
