Timely selection of optimal seed viruses with matched antigenicity between vaccine antigen and circulating viruses and with high yield is crucial for influenza vaccine efficacy and supply. Current methods are labor-intensive. This paper introduces MAIVESS, a machine-learning framework that streamlines this selection process by using molecular signatures of antigenicity and yield directly from clinical samples. Applied to A(H1N1)pdm09, MAIVESS identified optimal candidates experimentally confirmed for antigenicity and yield, potentially reducing selection time from months to days.

Cheng Gao, Feng Wen, Minhui Guan, Bijaya Hatuwal, Lei Li, Beatriz Praena, Cynthia Y. Tang, Jieze Zhang, Feng Luo, Hang Xie, Richard Webby, Yizhi Jane Tao, Xiu-Feng Wan