Programming languages are created by humans and the training dataset is complete enough to train LLMs with good results. Most importantly, natural language is the native domain of the programming code.

Whereas in biology, the natural domain is in physical/chemical/biological reactions occuring between organisms and molecules. The laws of interactions are not created by human, but by Creator(tm), and so the training dataset is barely capturing a tiny fraction and richness of the domain and its interactions. Because of this, any model will be inadequate