A Structure-Based Deep Learning Framework for Correcting Marine Natural Products’ Misannotations Attributed to Host–Microbe Symbiosis

Tian, Xiaohe; Lyu, Chuanyu; Zhou, Yiran; Zhang, Liangren; Fan, Aili; Liu, Zhenming

doi:10.3390/md24010020

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

A Structure-Based Deep Learning Framework for Correcting Marine Natural Products’ Misannotations Attributed to Host–Microbe Symbiosis

by

Xiaohe Tian

^†,

Chuanyu Lyu

^†

,

Yiran Zhou

,

Liangren Zhang

,

Aili Fan

and

Zhenming Liu

^*

State Key Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences, Peking University, Beijing 100191, China

^*

Author to whom correspondence should be addressed.

^†

These authors contributed equally to this work.

Mar. Drugs 2026, 24(1), 20; https://doi.org/10.3390/md24010020 (registering DOI)

Submission received: 4 December 2025 / Revised: 28 December 2025 / Accepted: 30 December 2025 / Published: 1 January 2026

(This article belongs to the Special Issue Chemoinformatics for Marine Drug Discovery)

Download Versions Notes

Abstract

Marine natural products (MNPs) are a diverse group of bioactive compounds with varied chemical structures, but their biological origins are often misannotated due to complex host–microbe symbiosis. Propagated through public databases, such errors hinder biosynthetic studies and AI-driven discovery. Here, we develop a structure-based workflow of origin classification and misannotation correction for marine datasets. Using CMNPD and NPAtlas compounds, we integrate a two-step cleaning strategy that detects label inconsistencies and filters structural outliers with a microbial-pretrained graph neural network. The optimized model achieves a balanced accuracy of 85.56% and identifies 3996 compounds whose predicted microbial origins contradict their Animalia labels. These putative symbiotic metabolites cluster within known high-risk taxa, and interpretability analysis reveal biologically coherent structural patterns. This framework provides a scalable quality-control approach for natural product databases and supports more accurate biosynthetic gene cluster (BGC) tracing, host selection, and AI-driven marine natural product discovery.

Keywords: marine natural products; origin classification; cheminformatics; host-microbe symbiosis; AI-driven drug discovery

Share and Cite

MDPI and ACS Style

Tian, X.; Lyu, C.; Zhou, Y.; Zhang, L.; Fan, A.; Liu, Z. A Structure-Based Deep Learning Framework for Correcting Marine Natural Products’ Misannotations Attributed to Host–Microbe Symbiosis. Mar. Drugs 2026, 24, 20. https://doi.org/10.3390/md24010020

AMA Style

Tian X, Lyu C, Zhou Y, Zhang L, Fan A, Liu Z. A Structure-Based Deep Learning Framework for Correcting Marine Natural Products’ Misannotations Attributed to Host–Microbe Symbiosis. Marine Drugs. 2026; 24(1):20. https://doi.org/10.3390/md24010020

Chicago/Turabian Style

Tian, Xiaohe, Chuanyu Lyu, Yiran Zhou, Liangren Zhang, Aili Fan, and Zhenming Liu. 2026. "A Structure-Based Deep Learning Framework for Correcting Marine Natural Products’ Misannotations Attributed to Host–Microbe Symbiosis" Marine Drugs 24, no. 1: 20. https://doi.org/10.3390/md24010020

APA Style

Tian, X., Lyu, C., Zhou, Y., Zhang, L., Fan, A., & Liu, Z. (2026). A Structure-Based Deep Learning Framework for Correcting Marine Natural Products’ Misannotations Attributed to Host–Microbe Symbiosis. Marine Drugs, 24(1), 20. https://doi.org/10.3390/md24010020

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.

Article Menu

A Structure-Based Deep Learning Framework for Correcting Marine Natural Products’ Misannotations Attributed to Host–Microbe Symbiosis

Abstract

Share and Cite

Article Metrics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI