Precise identification and quantification of amino acids is crucial for many biological applications. Here we report a copper(II)-functionalized Mycobacterium smegmatis porin A (MspA) nanopore with the N91H substitution, which enables direct identification of all 20 proteinogenic amino acids when combined with a machine-learning algorithm. The validation accuracy reaches 99.1%, with 30.9% signal recovery. The feasibility of ultrasensitive quantification of amino acids was also demonstrated at the nanomolar range. Furthermore, the capability of this system for real-time analyses of two representative post-translational modifications (PTMs), one unnatural amino acid and ten synthetic peptides using exopeptidases, including clinically relevant peptides associated with Alzheimer’s disease and cancer neoantigens, was demonstrated. Notably, our strategy successfully distinguishes peptides with only one amino acid difference from the hydrolysate and provides the possibility to infer the peptide sequence.

氨基酸的精确鉴定和定量对于许多生物应用至关重要。在这里，我们报道了一种带有 N91H 取代的铜 (II) 功能化耻垢分枝杆菌孔蛋白 A (MspA) 纳米孔，与机器学习算法结合使用时，可以直接识别所有 20 种蛋白氨基酸。验证准确率达到99.1%，信号恢复率为30.9%。还证明了在纳摩尔范围内超灵敏定量氨基酸的可行性。此外，该系统还能够使用外肽酶实时分析两种代表性的翻译后修饰 (PTM)、一种非天然氨基酸和十种合成肽，包括与阿尔茨海默病和癌症新抗原相关的临床相关肽。值得注意的是，我们的策略成功地区分出与水解产物仅具有一个氨基酸差异的肽，并提供了推断肽序列的可能性。