本文为原文前 6,000 字符的节选翻译，完整内容请查看原文。

Accelerating the search for antimicrobial peptides with Google Antigravity

October 15, 2026 Science. Accelerating the search for antimicrobial peptides with Google Antigravity. Pengfei Tian, Maxwell L. Bileschi, Zoi Tsangalidou, Sukhdeep Singh and Lucy J. Colwell. Explore scientific workflows in Google Antigravity. Share. New antibiotics are urgently needed to help with rising levels of antimicrobial resistance, but searching for new therapeutics is often a slow, labor-intensive process. With Google Antigravity, our agentic development platform, this process can now be made easier.

2026年10月15日，《科学》。利用 Google Antigravity 加速抗菌肽的搜索。Pengfei Tian, Maxwell L. Bileschi, Zoi Tsangalidou, Sukhdeep Singh 和 Lucy J. Colwell。探索 Google Antigravity 中的科学工作流程。分享。目前迫切需要新的抗生素来应对日益严重的抗菌素耐药性问题，但寻找新的治疗药物往往是一个缓慢且耗费人力的过程。借助我们的智能体开发平台 Google Antigravity，这一过程现在可以变得更加简便。

Google Antigravity provides an agentic science workbench that empowers biologists to discover, structurally evaluate, and prioritize promising targets quickly, compressing what used to be a fragmented search over multiple databases into a single session. It also helps researchers scan through millions of previously uncharacterized proteins in the UniProt database using ProtNLM, a model that predicts biological function from amino acid sequences.

Google Antigravity 提供了一个智能科学工作台，使生物学家能够快速发现、进行结构评估并确定有前景的靶点优先级，将以往分散在多个数据库中的搜索过程压缩到单一会话中。它还通过 ProtNLM（一种根据氨基酸序列预测生物功能的模型）帮助研究人员扫描 UniProt 数据库中数百万个此前未被表征的蛋白质。

This technical walkthrough shows how biologists used Google Antigravity to uncover a novel antimicrobial peptide candidate which has predicted activity against the pathogens involved in respiratory tract infections. The lead identified here is now undergoing wet-lab validation. Finding new antimicrobial peptides. Rising antimicrobial resistance is one of the most pressing global health crises of our time and could result in an estimated 10 million annual deaths globally by 2050. To prevent this, we urgently need to discover novel therapeutics.

本技术指南展示了生物学家如何利用 Google Antigravity 发现一种新型抗菌肽候选物，该候选物被预测对呼吸道感染相关的病原体具有活性。此处确定的先导化合物目前正在进行湿实验验证。寻找新的抗菌肽。日益严重的抗菌素耐药性是我们这个时代最紧迫的全球健康危机之一，到 2050 年，每年可能导致全球约 1000 万人死亡。为了防止这种情况，我们迫切需要发现新的治疗药物。

Antimicrobial peptides (AMPs), particularly bacteriocins, offer a promising alternative to traditional antibiotics because they have naturally evolved to target fundamental vulnerabilities in competing bacteria, from cell envelopes to essential intracellular machinery, making it difficult for pathogens to develop resistance. The exponential growth of genomic and metagenomic sequencing has populated global databases with billions of natural amino acid sequences. This pool of natural data has already produced globally used bacteriocins like Nisin, alongside recent genome-mined breakthroughs such as Darobactin, Dynobactin, and Lariocidin.

抗菌肽（AMPs），特别是细菌素，为传统抗生素提供了一种有前景的替代方案，因为它们在自然进化中能够靶向竞争细菌的基本弱点，从细胞包膜到重要的细胞内机制，这使得病原体难以产生耐药性。基因组和宏基因组测序的指数级增长已使全球数据库中充斥着数十亿条天然氨基酸序列。这一天然数据池已经产生了像乳酸链球菌素（Nisin）这样全球通用的细菌素，以及近期通过基因组挖掘取得的突破性成果，如 Darobactin、Dynobactin 和 Lariocidin。

Many, however, remain locked within the unannotated “dark matter” of protein databases, with their function uncharacterized because they aren’t similar to known proteins. For example, the world’s leading public resource of protein sequence and functional information, UniProt, holds millions of sequences with unknown functionality, transferred directly from genome sequencing projects. To bridge this gap, Google developed ProtNLM (Protein Natural Language Model), a deep learning model trained to directly read the underlying evolutionary “grammar” of amino acid sequences.

然而，许多蛋白质仍然被锁定在蛋白质数据库未注释的“暗物质”中，由于它们与已知蛋白质不相似，其功能尚未被表征。例如，全球领先的蛋白质序列和功能信息公共资源 UniProt 拥有数百万条功能未知的序列，这些序列直接来自基因组测序项目。为了弥补这一差距，Google 开发了 ProtNLM（蛋白质自然语言模型），这是一种经过训练的深度学习模型，可以直接读取氨基酸序列潜在的进化“语法”。

This allows the model to predict biological function directly from raw protein sequences, enriching millions of previously unnamed UniProt entries with high-confidence functional labels. Overview of the ProtNLM pipeline translating raw amino acid sequences into natural-language biological annotations for unreviewed UniProtKB/TrEMBL entries. The functional annotation scope of the system covers protein names, function comments, subcellular locations, and GO terms, in addition to keywords.

这使得该模型能够直接从原始蛋白质序列中预测生物功能，为数百万个此前未命名的 UniProt 条目提供了高置信度的功能标签。ProtNLM 流水线概述：将原始氨基酸序列转换为未经审查的 UniProtKB/TrEMBL 条目的自然语言生物学注释。该系统的功能注释范围涵盖蛋白质名称、功能注释、亚细胞位置和 GO 术语，以及关键词。

Collaborative intelligence. Scientific discovery is rarely a single step process; finding viable candidate proteins requires human intuition and iteration to incorporate new empirical evidence. Working in Antigravity, the researcher and AI agent begin by setting a focused biological goal: find and computationally validate novel AMPs in Streptococcus pseudopneumoniae. During an interactive, multi-turn session the researcher steers the scientific strategy, deciding which questions to ask and what evidence to collect, while the agent handles the heavy lifting of cross-database orchestration, structural analysis, and literature synthesis.

协作智能。科学发现很少是一个单一的步骤；寻找可行的候选蛋白质需要人类的直觉和迭代来整合新的经验证据。在 Antigravity 中工作时，研究人员和 AI 智能体首先设定一个明确的生物学目标：在假肺炎链球菌（Streptococcus pseudopneumoniae）中寻找并计算验证新型抗菌肽。在交互式的多轮会话中，研究人员引导科学策略，决定提出哪些问题以及收集哪些证据，而智能体则负责处理跨数据库编排、结构分析和文献综合等繁重工作。

Slide 1 of 6. Navigating a vast search space: The agent queries the database and filters millions of entries down to a handful of candidates worthy of attention. Here, the agent finds a bacteriocin (Uniprot ID: A0A0U0D591) with a length of 97 amino acids from the Lactococcin 972 family. Validating the novelty of a finding: To verify the novelty of this discovery, the researcher guides the agent to interrogate the candidate against curated knowledge in Swiss-Prot via the EMBL-EBI NCBI-BLAST service and confirm this peptide has no meaningful sequence homolog among manually reviewed proteins.

第 1 张（共 6 张）。导航广阔的搜索空间：智能体查询数据库并将数百万个条目过滤为少数值得关注的候选者。在这里，智能体发现了一种来自乳酸菌素 972（Lactococcin 972）家族的细菌素（Uniprot ID: A0A0U0D591），长度为 97 个氨基酸。验证发现的新颖性：为了验证这一发现的新颖性，研究人员引导智能体通过 EMBL-EBI NCBI-BLAST 服务针对 Swiss-Prot 中的精选知识对候选者进行查询，并确认该肽在人工审查的蛋白质中没有有意义的序列同源物。

Reasoning over 3D protein structure: The researcher steers the analysis towards structural evidence. The agent retrieves the candidate’s 3D model from the AlphaFold Protein Structure Database, querying it against experimentally determined structures to identify remote homologs, and superimposing the models to assess similarity. Here, the structural evidence corroborates ProtNLM’s prediction by showing the candidate’s fold closely matches an experimentally determined bacteriocin. The structure viewer in Antigravity allows this structural analysis to be seamlessly performed in one place.

对蛋白质 3D 结构进行推理：研究人员将分析导向结构证据。智能体从 AlphaFold 蛋白质结构数据库中检索候选者的 3D 模型，将其与实验确定的结构进行比对以识别远程同源物，并叠加模型以评估相似性。在这里，结构证据通过显示候选者的折叠与实验确定的细菌素高度匹配，证实了 ProtNLM 的预测。Antigravity 中的结构查看器允许在一个地方无缝执行此结构分析。

Reading the biological context around a gene: The agent looks beyond the molecule itself to its genomic neighborhood, reasoning over whether the surrounding machinery supports the AMP hypothesis. Inspecting the flanking genes, it spots a revealing pattern: our previously uncharacterized candidate sits at the upstream of a tightly coupled operon, just 6 base pairs ahead of a ABC transporter permease and its potential partner ATP-binding protein, alongside another annotated bacteriocin and an XRE transcriptional regulator. This provides strong contextual evidence that it is part of a working bacteriocin system.

读取基因周围的生物学背景：智能体不仅关注分子本身，还关注其基因组邻域，推理周围的机制是否支持抗菌肽假设。检查侧翼基因时，它发现了一个具有启发性的模式：我们此前未表征的候选者位于一个紧密耦合操纵子的上游，仅在 ABC 转运蛋白通透酶及其潜在的伴侣 ATP 结合蛋白之前 6 个碱基对处，旁边还有另一个已注释的细菌素和一个 XRE 转录调节因子。这提供了强有力的背景证据，表明它是工作中的细菌素系统的一部分。

Building a mechanistic hypothesis from the literature: By reviewing the existing literature, the agent proposes how the molecule might act and what pathogens it might act on. In this case, existing evidence suggests a mechanism of action targeting bacterial cell division, and a predicted potency spectrum spanning several drug-resistant pathogens. Generati

从文献中建立机制假设：通过回顾现有文献，智能体提出了该分子可能的作用方式以及它可能作用的病原体。在这种情况下，现有证据表明其作用机制是针对细菌细胞分裂，并且预测的效力范围涵盖了几种耐药病原体。Generati