miProbe: microbial peptide repository of biological embeddings

Microbial peptides represent a vast yet largely underexplored component of microbial dark matter, offering a valuable reservoir of bioactive molecules. Artificial Intelligence (AI) has revolutionized peptide mining, demonstrating high-throughput screening capabilities in tasks such as functional prediction and de novo design. To fully unlock the potential of microbial resource, we present miProbe (https://www.biosino.org/iMAC/miProbe), an integrative platform hosting 1.08 × 109 microbial peptides from disparate habitats. Uniquely, miProbe bridges the gap between raw sequences and AI applications by providing 8 TB of precomputed protein language model embeddings (e.g., ESM2-650M and ProtT5-XL-U50). Our analysis reveals substantial variation in peptide diversity and functional properties across habitats and microbial taxonomic groups, and highlights the utility of large-scale embedding resources in functional peptide mining. Equipped with robust APIs for seamless model integration, miProbe establishes a diverse and scalable data resource to accelerate AI-driven exploration of microbial dark matter and the rational design of novel therapeutics. Authors present miProbe, a platform aimed to consolidate peptide data and functional characterisation, and presenting a resource to accelerate AI-driven exploration of microbial dark matter.

Authors

Institutions

Publication Details

Journal
Nature Communications
Published
2026-11-13
DOI
https://doi.org/10.1038/s41467-026-79455-z
Primary Topic
Microbial Natural Products and Biosynthesis
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

miProbe: microbial peptide repository of biological embeddings

Dingfeng Wu, Xiaowen Pu, Lei Liu, Na Jiao et al.
1 links
Nature Communications
Microbial Natural Products and Biosynthesis
article

miProbe: microbial peptide repository of biological embeddings

Dingfeng Wu, Xiaowen Pu, Lei Liu, Na Jiao, Guoqing Zhang, Ruixin Zhu, Peigen Yao, Wei Wang, Qian Long, Bin Yang
article en

Abstract

Microbial peptides represent a vast yet largely underexplored component of microbial dark matter, offering a valuable reservoir of bioactive molecules. Artificial Intelligence (AI) has revolutionized peptide mining, demonstrating high-throughput screening capabilities in tasks such as functional prediction and de novo design. To fully unlock the potential of microbial resource, we present miProbe (https://www.biosino.org/iMAC/miProbe), an integrative platform hosting 1.08 × 109 microbial peptides from disparate habitats. Uniquely, miProbe bridges the gap between raw sequences and AI applications by providing 8 TB of precomputed protein language model embeddings (e.g., ESM2-650M and ProtT5-XL-U50). Our analysis reveals substantial variation in peptide diversity and functional properties across habitats and microbial taxonomic groups, and highlights the utility of large-scale embedding resources in functional peptide mining. Equipped with robust APIs for seamless model integration, miProbe establishes a diverse and scalable data resource to accelerate AI-driven exploration of microbial dark matter and the rational design of novel therapeutics. Authors present miProbe, a platform aimed to consolidate peptide data and functional characterisation, and presenting a resource to accelerate AI-driven exploration of microbial dark matter.

Nature Communications
Tongji University (CN), Fudan University (CN), Shanghai Institute of Nutrition and Health (CN), Shanghai First Maternity and Infant Hospital (CN)
National Natural Science Foundation of China, China Postdoctoral Science Foundation, National Key Research and Development Program of China
Microbial Natural Products and Biosynthesis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon