Targeting fish betanodavirus with BoltzGen cyclic-minipeptides

Cyclic-minipeptide candidates were computationally co-generated by deep-learning BoltzGen using home-designed Colab A100 notebooks against nonenveloped Nervous Necrosis Virus (NNV), an important worldwide expanded fish-farming pathogen. The Alphafold2-modeled capsid protein of grouper NNV was targeted by 6-, 13-and 21-amino acid sequences to co-generate CP // cyclic-minipeptide-conformers. The co-generated candidates facilitate recombinant/chemical synthesis, reduce physiological exoprotease degradation and conformer stability, increase environmental safety and would require improbable multiple-mutations to induce resistances. Despite exploring minimal percentages of the natural minipeptide wide-space, individually or in combination, these candidates constitute a novel proof-of-concept for experimentally-guided further explorations to control betanodavirus infections in real fish farming.