High-Affinity\nPeptides for Target Protein Screened\nin Ultralarge Virtual Libraries

High-throughput virtual screening (HTVS) has emerged\nas a pivotal\nstrategy for identifying high-affinity peptides targeting functional\nproteins, which are crucial for diagnostic and therapeutic applications.\nIn the HTVS of peptides, expanding the library capacity to enhance\npeptide sequence diversity, thereby screening out excellent affinity\npeptide candidates, remains a significant challenge. This study presents\na <i>de novo</i> design strategy that leverages directed\nmutation driven HTVS to evolve vast virtual libraries and screen peptides\nwith ultrahigh affinities for various target proteins. Utilizing a\ncomputer-generated library of 10<sup>4</sup> random 15-mer peptide\nscaffolds, we employed a self-developed algorithm for parallelized\nHTVS with Autodock Vina. The top 1% of designs underwent random mutations\nat a rate of 20% for six generations, theoretically expanding the\nlibrary to 10<sup>14</sup> members. This approach was applied to various\nprotein targets, including a tumor marker (alpha fetoprotein, AFP)\nand virus surface proteins (SARS-CoV-2 RBD and norovirus P-domain).\nStarting from the same 10<sup>4</sup> random 15-mer peptide library,\n peptides with high affinities in the nanomolar range for three protein\ntargets were successfully identified. The energy-saving and high-efficient\ndesign strategy presents new opportunities for the cost-effective\ndevelopment of more effective high-affinity peptides for various environmental\nand health applications.