Creative Proteomics integrate deep proteomics, phosphoproteomics, genomics, transcriptomics, and optional metabolomics to reveal the full molecular landscape of your disease. Our end-to-end, cohort-scale solution that translates multi-omics data into actionable disease subtypes and translational insights.
What is Disease Subtyping
Precision medicine relies on a deep, accurate view of the molecular differences driving disease. Conventional clinical classifications often lump together patients whose conditions differ at the biological level. This can weaken treatment effectiveness, hide important prognostic clues, and contribute to high drug development failure rates.
In contrast, multi-dimensional molecular profiling offers a more precise picture, especially when mass spectrometry-based proteomics is integrated with genomics and transcriptomics in a proteogenomics framework. It captures the functional state of tissues, identifies distinct patient subgroups, and enables the discovery of reliable prognostic biomarkers and promising therapeutic targets.
Technical Workflow for Multi-Omics Disease Subtyping
Technical Advantages of Our Multi-Omics Disease Subtyping Service
- High-resolution LC-MS/MS methods and optimized sample prep provide deep proteome and phosphoproteome coverage essential for subtype discrimination.
- Integrated analyses combining WES/WGS, RNA-seq, proteomics, and PTM layers to identify only discernible signals.
- Standardized SOPs, spike-in controls, randomized sample processing, and rigorous QC thresholds reduce technical artifacts and batch effects.
- Results are delivered as prioritized, evidence-scored biomarker and target lists with recommended validation paths.
Multi-Omics Analysis Service We Provide
Our service combines MS-based deep proteomics and phosphoproteomics with whole-exome/whole-genome sequencing (WES/WGS), transcriptomics (RNA-seq), and optional metabolomics or other PTM panels to generate a multi-layered view of tumor and disease biology. By profiling cohorts at scale, we quantify inter-individual molecular heterogeneity and use robust statistical and machine-learning approaches to group patients into biologically subtypes.
Recommended
- WES (Whole Exome Sequencing): variant discovery and copy number.
- RNA-seq (transcriptome): expression quantification and fusion/splicing detection.
- Proteomics (deep MS): global protein abundance profiling.
- Phosphoproteomics: signaling activity and pathway activation mapping.
Optional
- Metabolomics for metabolic phenotype and pathway flux insight.
- Targeted PTM assays (glycoproteomics, acetylomics) for focused hypotheses.
Data Analysis and Interpretation
- Clinical Information Statistics: Comprehensive statistical summarization and quality assessment of patient data, clinical characteristics, and treatment history.
- Feature Weighting and Combination Screening: Evaluation of the relative contribution of molecular features, followed by combinatorial screening to identify optimal feature sets.
- Biomarker Expression Profiling: Quantitative assessment of candidate biomarkers across different disease subtypes or patient groups.
- Biomarker Model Construction, Evaluation, and Comparison: Development of predictive models using selected biomarkers, with performance evaluation through metrics such as ROC curves, AUC, sensitivity, and specificity, and comparison across different models.
Application for Multi-Omics Disease Subtyping
- Oncology: tumor subtyping (e.g., immune hot vs cold, basal vs luminal), resistance mechanisms, neoantigen discovery, companion diagnostic biomarker nomination.
- Autoimmune and inflammatory diseases: tissue-based molecular endotypes linked to therapy response.
- Neurological disorders: where post-mortem or surgical tissue is available, PTM signaling may illuminate disease pathways.
- Infectious diseases: tissue or fluid proteomics to identify host-response subtypes.
Sample Requirements (Recommended)
| Sample | Requirement |
|---|---|
| Sample Type | Tumor tissue and matched adjacent normal tissue, or other disease-relevant samples |
| Sample Number | ≥ 100 paired samples |
| Sample Size | ≥ 200 mg per tissue sample |
| Preservation Method | Snap-frozen in liquid nitrogen and stored at −80°C until shipment |
| Shipping Conditions | Ship on dry ice to maintain sample integrity |
Why Choose Creative Proteomics
- Two decades of proteomics and translational research expertise with a publication track record.
- End-to-end capability from sample handling to clinical translation.
- Flexible, modular packages to match scientific goals and budgets.
- Transparent reporting, reproducible pipelines, and regulatory-aware deliverables.
- A dedicated scientific team available for study design, interpretation, and downstream validation planning.
References
- Mani D R, et al. Cancer proteogenomics: current impact and future prospects. Nature Reviews Cancer, 2022, 22(5): 298-313.
- Mertins P, et al. Reproducible workflow for multiplexed deep-scale proteome and phosphoproteome analysis of tumor tissues by liquid chromatography–mass spectrometry. Nature protocols, 2018, 13(7): 1632-1661.
- Ang M Y, et al. Proteogenomics: from next-generation sequencing (NGS) and mass spectrometry-based proteomics to precision medicine. Clinica chimica acta, 2019, 498: 38-46.
Proteogenomic analysis of human colon cancer reveals new therapeutic opportunities
Journal: Cell
Impact factor: 42.5
Published: 2019
DOI: 10.1016/j.cell.2019.03.030
Backgrounds
Colorectal cancer (CRC) is molecularly heterogeneous and remains a leading cause of cancer mortality worldwide. Prior large-scale genomic and transcriptomic studies (e.g., TCGA, consensus molecular subtypes) revealed important molecular classes but have not fully delivered clinically actionable biomarkers or therapeutic targets. The Clinical Proteomic Tumor Analysis Consortium (CPTAC) aimed to add proteomic and phosphoproteomic layers to genomic/transcriptomic profiling to capture functional and signaling changes that genomes/transcriptomes alone can miss.
Materials & Methods
Specimens from 110 colon cancer patients (tumor, matched normal adjacent tissue, blood)
Whole-exome sequencing (WXS), copy number arrays, RNA-seq, miRNA-seq, label-free shotgun proteomics, and TMT-based global plus phosphoproteomics.
| WXS | Sequencing of tumor and matched normal DNA to identify somatic mutations, CNVs, and MSI status. | APC mutations in 83/110 samples; MSI-high in 16% of tumors. |
| RNA-seq | Transcriptome profiling to measure gene expression, detect fusions, and infer immune composition. | >2,300 genes significantly altered (FDR < 0.05) between tumor and NAT. |
| Global Proteomics (TMT-based) | Multiplexed LC-MS/MS to quantify protein abundance across tumor and NAT samples. | ~9,800 proteins quantified in >80% of samples. |
| Phosphoproteomics | Enrichment and LC-MS/MS analysis of phosphopeptides to study kinase signaling. | >27,000 phosphosites identified; 2,100 significantly altered (FDR < 0.05). |
| Proteogenomic Integration | Cross-referencing genomic, transcriptomic, proteomic, and phosphoproteomic data. | Immune-active subtype showed 5-year survival rate ~82% vs ~56% in proliferative subtype. |
Results
- Produced a catalog of colon cancer–associated proteins and phosphosites (tumor vs adjacent normal) and observed protein-level events not predicted by RNA/genome alone.
- Identified molecular substructure grouping tumors into MSI (microsatellite instability), CIN (chromosomal instability), and Mesenchymal-like subtypes using multi-omics clustering — providing a unified view complementary to existing CMS classifications. Proteomics revealed subtype-specific biology.
- Phosphoproteomics implicated Rb phosphorylation (increased phospho-Rb) as a driver of proliferation. It decreased apoptosis in some colon cancers — suggesting CDK inhibition (targeting kinases upstream of Rb phosphorylation) as a therapeutic hypothesis where RB1 mutation is absent.
- Proteomics linked reduced CD8+ T-cell infiltration with increased glycolysis in MSI-high tumors, proposing glycolysis inhibition as one strategy to sensitize MSI tumors to immune checkpoint blockade.
- Proteomics supported identifying tumor antigens and neoantigen candidates in many tumors, informing immunotherapy and vaccine strategies.
Fig. 1 Schematic overview of the study.
Conclusions
Integrating proteomic and phosphoproteomic data with genomic and transcriptomic profiles reveals functional tumor biology and therapeutic vulnerabilities not visible from DNA/RNA alone. The study produced prioritized, evidence-scored hypotheses (e.g., targeting phospho-Rb via CDKs, combining glycolysis inhibitors with immunotherapy for MSI tumors) and public multi-omics datasets to support further validation and translational work. The paper demonstrates how proteogenomic multi-omics can sharpen disease subtyping and accelerate precision-medicine target nomination.
How does multi-omics analysis improve disease subtyping?
Multi-omics combines data from genes, RNA, proteins, and metabolites to uncover differences missed by single-omics. For example, proteomics can reveal excess receptors or protein modifications invisible in genomic data, improving target discovery.
How does multi-omics contribute to understanding complex and co-morbid diseases?
Complex diseases often have overlapping conditions and are best understood through long-term multi-omics profiling. This approach reveals disease patterns, connections, and key drivers, supporting precise classification and personalized treatment.
Can multi-omics help identify mechanisms of therapeutic resistance?
Yes, integrated proteomics and phosphoproteomics are particularly powerful for revealing activated bypass pathways, compensatory signaling, and post-translational mechanisms underlying resistance, which can inform combination strategies.
Can FFPE samples be used for multi-omics profiling?
FFPE samples work with specialized workflows, such as targeted proteomics and optimized nucleic acid extraction kits. However, degraded RNA and crosslinked proteins may reduce sensitivity. Plan ahead by discussing FFPE-specific methods and potential trade-offs with us.
Demo: A proteogenomics data-driven knowledge base of human cancer
Fig 2. Visualizations to facilitate efficient exploration and reasoning of pan-cancer, multi-omics data (Liao Y X, et al., 2023).
Fig 3. Proteogenomics insights into clinical phenotypes (Liao Y X, et al., 2023).