Core Insights - On October 8, 2025, 15 papers were published in the prestigious journal Nature, with 4 authored by Chinese scholars [2][4][6][8] Group 1: Research Contributions - Professor Zhao Huimin from the University of Illinois Urbana-Champaign published a paper titled "Enzyme specificity prediction using cross attention graph neural networks," introducing an innovative AI model called EZSpecificity that predicts enzyme substrate specificity with high accuracy, aiding in enzyme engineering and synthetic biology [2] - Professor Zhou Peng from Fudan University published a paper titled "A full-featured 2D flash chip enabled by system integration I," focusing on advancements in 2D flash memory technology through system integration [4] - MIT's Liu Qi co-authored a paper titled "Quantum-amplified global-phase spectroscopy on an optical clock transition," contributing to the field of quantum spectroscopy [6] - Professors Lou Zhenkun and Huang Jinzhu from Mayo Clinic published a paper titled "KCTD10 is a sensor for co-directional transcription–replication conflicts," revealing a ubiquitin ligase complex that detects transcription-replication conflicts and regulates replication body progression to prevent DNA damage [8]

撰文丨王聪 编辑丨王多鱼 排版丨水成文 酶 （ Enzyme ） 是生命的分子机器，其功能的一个关键特性是底物特异性——即酶能够识别并选择性地作用于特定底物的能力。这种特异性源自酶活性位点的三 维 （3D） 结构以及反应复杂的过渡态。实际上，许多酶并非传统认为的那样"专一"，它们可能具有多功能性，除了作用于其最初进化时涉及的底物，还能作用 于其他结构相似的底物，从而催化不同类型的反应。 此外，数以百万计的 已知的酶仍缺乏可靠的底物特异性信息，这阻碍了它们的实际应用以及我们对自然界生物催化多样性的全面理解。 2025 年 10 月 8 日， 伊利诺伊大学厄巴纳-香槟分校 （UIUC） 赵惠民 教授团队 （ 崔海洋 、 苏雨峰 为共同第一作者 ） 在国际顶尖学术期刊 Nature 上发表 了 题为： Enzyme specificity prediction using cross attention graph neural networks 的研究论文，论文第一作者 崔海洋 博士现为 南京师范大学教授。 该研究提出了一种结合 跨注意力机制 （Cross-Attention） 与 SE(3)-等变图神经网络 （ ...