Progress of the Chinese Multi-omics Advances In Sepsis (CMAISE) database: a review

Introduction to the Chinese Multi-omics Advances In Sepsis (CMAISE) database

Sepsis, a life-threatening organ dysfunction caused by infection, carries high mortality globally. The Surviving Sepsis Campaign study of nearly 30,000 patients with severe sepsis or septic shock reported an overall hospital mortality rate of 32.8% (1). Furthermore, survivors frequently face significant long-term morbidity, including frequent hospital readmissions, reduced quality of life, and substantial cognitive and functional impairments, often necessitating ongoing specialized care (2). Despite advancements in critical care, the pathophysiology of sepsis remains complex, involving dysregulated immune responses, multi-organ involvement, and molecular changes across multiple omics (such as genomics, proteomics, and metabolomics) levels, compounded by high patient heterogeneity. This urgently calls for innovative research approaches to enhance diagnosis and treatment (3,4).

Currently, several medical databases are available for sepsis research (5-8), but they suffer from limitations such as singular time data, inadequate technological approaches, lack of multi-omics integration, and single-source data origins. To address these challenges, the CMAISE database (https://github.com/zh-zhang1984/CMAISE/wiki) was launched in 2020. Led by Dr. Zhongheng Zhang, this multicenter, multi-omics research project aims to elucidate the molecular mechanisms and clinical characteristics of sepsis by integrating clinical data and multi-omics data from multiple medical centers. Its goal is to uncover biomarkers and potential therapeutic targets for sepsis through multidimensional data analysis, with the aim of improving early diagnosis rates, optimizing treatment strategies, and enhancing patient clinical outcomes. By establishing a robust, publicly accessible database, CMAISE supports collaborative research and accelerates progress in precision medicine for sepsis.

This paper aims to provide a comprehensive overview of the CMAISE database’s development, key achievements, and future directions in advancing sepsis research. By detailing the database’s structure, data quality, and significant findings—such as novel biomarkers and personalized treatment strategies—it highlights CMAISE’s role in bridging multi-omics research with clinical practice. The report also outlines ongoing efforts to expand the database and foster global collaboration, offering a valuable resource for researchers and clinicians in emergency and critical care medicine to improve sepsis diagnosis, management, and patient outcomes.

Data characteristics

Currently, the data from CMAISE-v1.0 has been officially released through the National Genomics Data Center (NGDC, https://ngdc.cncb.ac.cn/), making it accessible and available for use by researchers worldwide. Meanwhile, data collection for CMAISE-v2.x is actively underway to further expand and enrich the database’s content.

• Patient count: the released CMAISE v1.0 includes clinical information and biospecimens from 597 septic patients across 35 medical centers, covering diverse clinical phenotypes such as early-stage sepsis, septic shock, and complications [e.g., acute kidney injury (AKI), acute respiratory distress syndrome].
• Patient types: the released CMAISE v1.0 includes clinical information and blood samples from sepsis patients, covering various clinical phenotypes such as early sepsis, septic shock, and complications (e.g., AKI, acute respiratory distress syndrome).
• Time dimension: the data encompasses samples from the 1st, 3rd, and 5th days of admission, providing a dynamic perspective on disease progression.
• Quality control: the CMAISE database ensures data scientific integrity and reliability through a comprehensive quality control system. Samples undergo assessments including optical density (O.D.) 260/280 and O.D. 260/230 (9,10), concentration, total quantity, and RNA integrity evaluation [RNA integrity number (RIN) value] using the Agilent 2100 Bioanalyzer (11). RNA samples are categorized into three quality tiers: “qualified”, “at-risk for library construction”, and “unqualified”. These classifications are determined according to specific project requirements, including those for microRNA sequencing and transcriptome sequencing. “Qualified” RNA signifies that the sample has sufficient quality, quantity, and integrity to proceed with experimental protocols. “At-risk for library construction” RNA indicates that while the RNA quality is acceptable, the quantity or integrity is somewhat below the standard threshold. This may result in library construction failure, abnormal insert fragments, a higher ribosomal ratio in sequencing, reduced library randomness, and lower data output. “Unqualified” RNA refers to samples that are either degraded or have severely inadequate quantities, and their use in experiments is not advised. The RNA quality control chromatogram is shown in Figure 1.

Figure 1 Example of RNA quality control chromatogram. nt (nucleotide): indicates the length of RNA in terms of the number of nucleotides. FU (fluorescence units): represents fluorescence intensity, which correlates with sample concentration. Marker: an RNA standard with a known length (25 nt) and concentration, added during quality control to determine the size distribution of sample fragments. S (sedimentation coefficient): used to differentiate ribosomal subunits based on their sedimentation rates during centrifugation. LM, low molecular weight RNA.

Technical platforms and data processing

CMAISE v1.0 employs advanced sequencing and mass spectrometry technologies to generate multi-omics data. For example, transcriptomics data is produced using the Illumina NovaSeq platform (12), proteomics data is generated using Olink (13) and DIA (14) technologies, and single-cell sequencing is conducted with GEXSCOPE technology (15) (Table 1).

Database storage and usage

In the global management of bioinformatics data, the submission, storage, and management of data are critical components. Currently, the CMAISE database’s data is stored at the NGDC, a core institution for the submission, storage, and management of global bioinformatics data. It provides a variety of data submission and storage services, ensuring the efficient management and utilization of bioinformatics data. In 2024, CMAISE made significant progress in database development, with the most notable milestone being the public release of CMAISE-v1.0. This version includes data from sepsis patients across 35 medical centers, under the project ID PRJCA006118 (https://ngdc.cncb.ac.cn/bioproject/browse/PRJCA006118). The project includes the following content, as detailed in Table 2. Additionally, research does not always need to start with raw data; downstream analysis can begin with expression matrices. The CMAISE database also provides RNA sequencing expression matrix data (https://ngdc.cncb.ac.cn/omix/release/OMIX006457) and single-cell RNA sequencing expression matrix data (https://ngdc.cncb.ac.cn/omix/release/OMIX005600). The database’s coding convention is that raw data containing base sequences starts with “HRA”, while analyzed expression matrices start with “OMIX”.

Significant achievements of CMAISE: advancements in precision medicine through artificial intelligence (AI)-driven research

The CMAISE database has transformed sepsis research by integrating multi-omics and clinical data to uncover molecular mechanisms and enhance treatment strategies, significantly impacting emergency and critical care. By combining transcriptomics, proteomics, and single-cell sequencing, CMAISE has advanced precision medicine through molecular signatures and dynamic classification frameworks. For example, a six-protein proteomic signature (area under the curve =0.802) optimizes restrictive fluid strategies in septic shock, reducing in-hospital mortality from 16% to 13% (16). Gene expression signatures predict AKI trajectories, enabling early interventions to reduce kidney damage (17). Single-cell analyses reveal diverse neutrophil subgroups tied to inflammation, immunosuppression, and tissue repair, informing targeted therapies (18). Continuous “disease axes” [e.g., shock and systemic inflammatory response syndrome (SIRS) axes] replace traditional subtypes, better capturing patient heterogeneity for tailored treatments (19). Studies on isoform switching and ulinastatin’s immunomodulatory effects, such as neutrophil regulation, identify novel regulatory layers and therapeutic targets (20,21). Deep learning integrating chest computed tomography (CT) and plasma proteomics further identifies image-expression axes linked to severe coronavirus disease 2019 (COVID-19) outcomes, with implications for sepsis management (22). These achievements collectively improve early diagnosis, personalize treatment, and enhance outcomes in critical care settings.

Researchers can utilize CMAISE’s open-access data, hosted at the NGDC (project ID PRJCA006118), to advance sepsis research and clinical practice. The database supports refined classification by integrating multi-omics profiles to redefine subtypes and identify biomarkers (19,20). It facilitates exploration of treatment modalities, such as optimized fluid strategies or immune-modulating therapies, using molecular and clinical data (16,21). Patient characteristics, tracked over days 1, 3, and 5, enable analysis of disease progression and outcome predictors (17,19). For international researchers, CMAISE’s data from 35 Chinese hospitals allows comparative studies on geographic and ethnic differences in sepsis, particularly when paired with global datasets (19). These applications yield practical tools, such as predictive models and biomarkers, enhancing early detection and personalized care in emergency and critical care worldwide.

Participation and collaboration

CMAISE is dedicated to fostering collaboration among Chinese researchers, with its database openly accessible via the NGDC. This transparency has sparked international partnerships and a variety of data-mining initiatives, all aimed at advancing sepsis research.

In 2024, the CMAISE team actively engaged the wider scientific community by organizing several symposiums. These efforts heightened the database’s profile and expanded its use in sepsis studies. The project champions multicenter and multidisciplinary cooperation, drawing in a growing number of medical researchers.

Following the launch of CMAISE-v1.0, the development of CMAISE-v2.x is progressing rapidly. As of March 31, 2025, 291 sepsis patients have been screened, with 160 included in the dataset, and an increasing number of medical centers are joining the CMAISE project. In addition to the standard data from CMAISE-v1.0, CMAISE-v2.x incorporates new elements like patient fecal samples and follow-up data. Details on how to participate can be found via the quick response (QR) code in Figure 2.

Figure 2 How to join the CMAISE-2.0, including participation QR code. CMAISE, Chinese Multi-omics Advances In Sepsis; EDC, electronic data capture; ICU, intensive care unit; QR, quick response; RNA-seq, RNA-sequencing.

Future outlook

Looking ahead to 2025 and beyond, CMAISE aims to amplify its capabilities and impact through the following initiatives:

• Integration of AI technology: the potential of AI in sepsis management is immense, and CMAISE plans to leverage this technology to drive advancements in critical care medicine. By harnessing AI, CMAISE intends to facilitate a transformative approach to understanding the complex dynamics of sepsis, aiming to bridge the gap between research and clinical practice. This strategic integration will empower the database to contribute to a new era of precision medicine, where data-driven insights can guide the development of innovative diagnostic tools and therapeutic strategies, ultimately improving patient outcomes on a global scale.
• Data expansion: high-quality, diverse data is the cornerstone of research and application. CMAISE will continue to gather and integrate additional multi-omics data types, including longitudinal samples and fecal samples from sepsis patients, to better track disease progression patterns. This expansion underscores CMAISE’s commitment to creating a holistic data framework that captures the intricate and dynamic nature of sepsis across various biological dimensions. By broadening its data scope, CMAISE aims to fuel groundbreaking research that can redefine how sepsis is understood and managed, fostering a global platform for innovation. This visionary approach is designed to support the development of novel strategies that address the challenges of sepsis, ultimately improving patient outcomes worldwide.
• Ethics and privacy protection: CMAISE upholds rigorous ethical reviews, data encryption, and access controls to protect patient privacy and ensure data use aligns with medical ethics, fostering trust within the global research community. This trust is built on the responsible handling of sensitive data. By sticking to the highest ethical standards, CMAISE sets a high bar for data integrity. The goal is clear: ensure that all resources are used in a way that respects patient rights and moves medical science forward responsibly. CMAISE’s commitment to ethics supports impactful research and earns public trust.
• Fostering collaboration: the project will seek to onboard more partners for data collection while integrating with existing databases to build a more comprehensive sepsis research resource, including partnerships with international consortia and research networks. CMAISE is all about bringing people together to tackle sepsis on a global scale. The idea is to break down the walls between institutions and countries, creating a single platform where everyone can work together. CMAISE amplifies research outcomes by fostering collaborative data integration. By building a network that’s open to all, CMAISE can tap into a wide range of viewpoints and expertise. When knowledge is shared freely, innovation happens naturally. Through collaboration, CMAISE aims to become a global leader in sepsis research. By working hand in hand, the project aims to speed up progress and make a real difference in the battle against this deadly condition.
• Clinical translation: efforts will focus on converting research findings into clinical tools, such as developing biomarkers for early sepsis detection and models to predict patient outcomes. CMAISE translates research discoveries into practical solutions to improve patient outcomes. By zeroing in on outcomes that doctors can act on, CMAISE wants to change how sepsis is managed. The goal is simple: give healthcare providers better tools to act faster and save more lives. CMAISE uses data to develop actionable tools, such as biomarkers for early detection and models to predict patient outcomes. These solutions are designed to fill critical gaps in critical care medicine and ensure that research doesn’t just sit on a shelf but gets used where it matters most.

Implementing these initiatives presents several challenges. AI integration may face difficulties in processing complex multi-omics data and ensuring models are interpretable for clinicians in fast-paced emergency settings. Data expansion could be hindered by inconsistent sample quality across centers and the high costs of advanced sequencing. Ethics and privacy efforts must navigate varying international regulations, risking delays in data sharing. Collaboration may encounter obstacles from institutional data-sharing restrictions or differing research priorities. Clinical translation requires rigorous validation of tools across diverse populations to ensure applicability in real-world critical care. To overcome these, CMAISE will prioritize user-friendly AI tools with clear documentation, implement uniform protocols for sample collection, adopt flexible encryption to comply with global standards, establish transparent data-sharing agreements, and conduct collaborative validation studies to ensure tools are robust and practical for emergency and critical care settings.

These plans will build on the strong foundation laid in 2024, propelling CMAISE toward becoming a globally leading resource for sepsis research. This will enhance its support for critical care studies and clinical practice, providing robust tools and platforms for early diagnosis, precision treatment, and improved patient outcomes in sepsis.

Conclusions

CMAISE marks a significant stride in the fight against sepsis, offering global researchers a rich, openly accessible resource. By combining clinical and multi-omics data, CMAISE has already yielded valuable insights, such as gene signatures for predicting AKI trajectories. As the project evolves, its potential to transform the understanding, diagnosis, and treatment of sepsis is vast. Through fostering collaboration and innovation, CMAISE ensures its impact reaches far beyond its originating institutions, contributing to worldwide efforts to mitigate the devastating toll of sepsis.

Acknowledgments

None.

Footnote

Peer Review File: Available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-2025-23/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jeccm.amegroups.com/article/view/10.21037/jeccm-2025-23/coif). Y.H. and Y.Y. serve as the Editors-in-Chief of Journal of Emergency and Critical Care Medicine. Z.Z. serves as the unpaid Executive Editor of Journal of Emergency and Critical Care Medicine. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Levy MM, Rhodes A, Phillips GS, et al. Surviving Sepsis Campaign: association between performance metrics and outcomes in a 7.5-year study. Intensive Care Med 2014;40:1623-33. [Crossref] [PubMed]
2. Cecconi M, Evans L, Levy M, et al. Sepsis and septic shock. Lancet 2018;392:75-87. [Crossref] [PubMed]
3. Hu B, Ji W, Bo L, et al. How to improve the care of septic patients following "Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021"? J Intensive Med 2023;3:144-6. [Crossref] [PubMed]
4. Zhang H, Dong N, Yao Y. Optimal strategy for treatment of sepsis based on the host inflammatory reaction and immune response. J Intensive Med 2024;4:175-80. [Crossref] [PubMed]
5. Block C, Engelhard D. Antibiotic prescribing practices in older children. Lancet 1983;1:868. [Crossref] [PubMed]
6. Hästbacka J, But A, Strandberg G, et al. Risk of malignant disease in 1-year sepsis survivors, a registry-based nationwide follow-up study. Crit Care 2023;27:376. [Crossref] [PubMed]
7. Johnson AE, Pollard TJ, Shen L, et al. MIMIC-III, a freely accessible critical care database. Sci Data 2016;3:160035. [Crossref] [PubMed]
8. Li Y, Tan R, Chen Y, et al. SC2sepsis: sepsis single-cell whole gene expression database. Database (Oxford) 2022;2022:baac061. [Crossref] [PubMed]
9. Shah SG, Rashid M, Verma T, et al. Establishing a correlation between RIN and A(260/280) along with the multivariate evaluation of factors affecting the quality of RNA in cryopreserved cancer bio-specimen. Cell Tissue Bank 2019;20:489-99. [Crossref] [PubMed]
10. Zeng Y, Tang X, Chen J, et al. Optimizing total RNA extraction method for human and mice samples. PeerJ 2024;12:e18072. [Crossref] [PubMed]
11. Senst A, Bonsiepe H, Kron S, et al. Application of the Agilent 2100 Bioanalyzer instrument as quality control for next-generation sequencing. J Forensic Sci 2024;69:2192-6. [Crossref] [PubMed]
12. Modi A, Vai S, Caramelli D, et al. The Illumina Sequencing Protocol and the NovaSeq 6000 System. Methods Mol Biol 2021;2242:15-42. [Crossref] [PubMed]
13. Petrera A, von Toerne C, Behler J, et al. Multiplatform Approach for Plasma Proteomics: Complementarity of Olink Proximity Extension Assay Technology to Mass Spectrometry-Based Protein Profiling. J Proteome Res 2021;20:751-62. [Crossref] [PubMed]
14. Lou R, Cao Y, Li S, et al. Benchmarking commonly used software suites and analysis workflows for DIA proteomics and phosphoproteomics. Nat Commun 2023;14:94. [Crossref] [PubMed]
15. Gao Q, Wu ZY, Li HY, et al. Preliminary Results of Single-cell Transcriptome Sequencing in Renal Arterial Lesions of Takayasu Arteritis. Zhongguo Yi Xue Ke Xue Yuan Xue Bao 2023;45:80-7. [Crossref] [PubMed]
16. Zhang Z, Chen L, Sun B, et al. Identifying septic shock subgroups to tailor fluid strategies through multi-omics integration. Nat Commun 2024;15:9028. [Crossref] [PubMed]
17. Zhang Z, Chen L, Liu H, et al. Gene signature for the prediction of the trajectories of sepsis-induced acute kidney injury. Crit Care 2022;26:398. [Crossref] [PubMed]
18. Hong Y, Chen L, Sun J, et al. Single-cell transcriptome profiling reveals heterogeneous neutrophils with prognostic values in sepsis. iScience 2022;25:105301. [Crossref] [PubMed]
19. Zhang Z, Chen L, Liu X, et al. Exploring disease axes as an alternative to distinct clusters for characterizing sepsis heterogeneity. Intensive Care Med 2023;49:1349-59. [Crossref] [PubMed]
20. Chen L, Chen K, Hong Y, et al. The landscape of isoform switches in sepsis: a multicenter cohort study. Sci Rep 2022;12:10276. [Crossref] [PubMed]
21. Chen L, Jin S, Yang M, et al. Integrated Single Cell and Bulk RNA-Seq Analysis Revealed Immunomodulatory Effects of Ulinastatin in Sepsis: A Multicenter Cohort Study. Front Immunol 2022;13:882774. [Crossref] [PubMed]
22. Hong Y, Chen L, Yu Y, et al. Deep learning integration of chest computed tomography and plasma proteomics to identify novel aspects of severe COVID-19 pneumonia. J Intensive Med 2025;5:252-61. [Crossref] [PubMed]