Scientists at the Qingdao Institute of Bioenergy and Bioprocess Technology (QIBEBT), Chinese Academy of Sciences (CAS) have developed an innovative computational tool, RamEx, which addresses the computational bottleneck faced in high-throughput microbial Ramanomics. This breakthrough, published in the Journal of Microbiome on February 10, 2026, promises to accelerate the analysis of massive spectral datasets generated by Raman flow cytometry, unlocking new opportunities for microbial profiling.

Ramanome has emerged as a powerful type of big-data for single-cell metabolic phenome, enabling label-free, non-invasive profiling of cell types and physiological states. With the advent of high-throughput Raman flow cytometry, generation of massive ramanomic datasets has created a significant "computational bottleneck." Existing analysis tools, often limited to basic preprocessing or ill-suited for the high-dimensional and noisy nature of ramanome datasets, struggle to translate this data deluge into biological insights.

RamEx overcomes these challenges by streamlining the entire ramanome analysis pipeline, from data preprocessing and automated quality control to advanced data mining. "The innovation of RamEx lies in its tailored approach to the unique characteristics of microbial spectral data," said ZHANG Yanmei, a graduate student from the Single-Cell Center at QIBEBT. "We developed the Iterative Convolutional Outlier Detection (ICOD) algorithm to address the challenge of noisy spectra. Unlike traditional methods that rely on manual, experience-based thresholds, ICOD works in an unsupervised manner to dynamically identify and remove spectral artifacts, ensuring high-quality data for downstream analysis.".

The platform's capability was rigorously validated using diverse datasets, including pathogenic bacteria, probiotics, and yeast fermentation processes. In particular, RamEx was able to capture phenotypic heterogeneity in isogenic yeast populations—cells with identical genotypes—by detecting subtle metabolic variations and tracing the dynamic accumulation of intracellular macromolecules, such as lipids, proteins, and nucleic acids.

"RamEx bridges the gap between high-throughput ramanome acquisition and computational analysis," stated Prof. SUN Luyang, who led the study. "It provides a robust toolkit that enables researchers to explore microbial ecology, metabolic interactions, and environmental stress responses at an unprecedented single-cell resolution."

"With this open-source framework, we aim to standardize ramanome analysis and facilitate the broader adoption of the ramanomics approach" added Prof. XU Jian from the Single-Cell Center, QIBEBT, who co-led the study. Together with the RamanAI system, RamEx is supporting the iMAPS (in-situ Metabolic Atlas Projects @ Single-cell) Consortium, which is building a real-time, in-situ, automated, and intelligent network of metabolic sensor, strain miner, and ecological restorer for microbiomes across the globe (iMAPS.info).

Software Availability: https://github.com/qibebt-bioinfo/RamEx

RamEX: a modularized pipeline for ramanome analysis

（Text by ZHANG Yanmei and Image by LIU Yang)