'Multi-Omics Integration'

'Combine DNA and RNA analysis for comprehensive cancer profiling'

Multi-Omics Cancer Analysis

Omics807 uniquely combines DNA variant analysis with single-cell RNA-seq analysis to provide comprehensive multi-omics cancer insights.

What is Multi-Omics?

Multi-omics integrates multiple layers of biological data:

  1. Genomics (DNA) - Somatic mutations from DeepSomatic
  2. Transcriptomics (RNA) - Gene expression from Cell2Sentence
  3. Integration - Cross-reference mutations with expression patterns

Why Multi-Omics Matters

DNA Analysis Alone

Shows what mutations exist: - TP53 mutation at chr17:7577538 - BRAF V600E mutation - KRAS G12D mutation

RNA Analysis Alone

Shows what genes are expressed: - High EGFR expression - Low PD-L1 expression - Activated immune cells

Multi-Omics Together

Shows how mutations affect biology: - TP53 mutation → MDM2 overexpression → MDM2 inhibitor sensitivity - BRAF V600E → MAPK pathway activation → targeted therapy response - High PD-L1 + T cell infiltration → immunotherapy candidate

Pathway-Level Multi-Omics Analysis 🆕

New in v1.1.0 - Omics807 now provides advanced pathway-level correlation to identify functional convergence between DNA variants and RNA expression, even when direct gene-level overlap is minimal.

Why Pathway Analysis?

Traditional multi-omics integration looks for gene-level overlap: - DNA: BRAF mutation detected - RNA: BRAF expression measured - Limitation: What if the mutated genes aren't highly expressed?

Pathway-level analysis identifies functional convergence: - DNA: BRAF V600E mutation affects MAPK signaling - RNA: MEK1, MEK2, ERK1, ERK2 (downstream MAPK pathway genes) are overexpressed - Insight: Even if BRAF RNA is low, the MAPK pathway is clearly activated

How It Works

  1. Gene-to-Pathway Mapping
  2. Maps DNA variants to KEGG and Reactome pathways
  3. Maps RNA expression changes to the same pathways

  4. Identifies convergent pathways from both data types

  5. Pathway Enrichment Scoring

  6. Calculates enrichment across 300+ cancer-related pathways
  7. Identifies statistically significant pathway alterations

  8. Prioritizes pathways with multi-omics evidence

  9. Druggable Target Identification

  10. Queries ChEMBL database for FDA-approved drugs
  11. Matches drugs to affected pathways
  12. Provides therapeutic recommendations with clinical evidence

Example Use Case

Patient Data: - DNA: TP53 mutation, PTEN deletion - RNA: PI3K pathway genes highly expressed (PIK3CA, AKT1, mTOR)

Traditional Analysis: - Gene overlap: None (TP53 and PTEN not in top expressed genes) - Result: Limited integration insights

Pathway Analysis: - Convergent pathway: PI3K-AKT signaling (KEGG:04151) - DNA evidence: PTEN deletion removes pathway inhibition - RNA evidence: Downstream PI3K-AKT genes overexpressed - Druggable target: PI3K inhibitor (Alpelisib, FDA-approved) - Clinical recommendation: Consider PI3K pathway targeting therapy

Benefits

✅ Identifies convergence when genes don't overlap
✅ Provides mechanistic insights (pathway dysregulation)
✅ Actionable druggable targets with FDA-approved drugs
✅ Concise, pathway-focused insights (no verbose text)
✅ Automated enrichment from KEGG, Reactome, and ChEMBL

Getting Started

Omics807's multi-omics integration is now available! You can:

1. Upload Both Data Types

  • DNA: BAM files (tumor + normal)
  • RNA: scRNA-seq expression matrix

2. Automatic Integration

Omics807 automatically: - Runs deep learning variant calling - Performs Cell2Sentence expression analysis - Enriches variants with population genetics, protein structures, drug targets, pathways - Cross-references mutations with expression - Generates unified insights with comprehensive clinical evidence

3. Integrated Dashboard

View combined results: - Mutation map with expression overlay - Pathway enrichment analysis - Treatment recommendations based on both layers

Use Cases

1. Precision Oncology

Patient Profile: - DNA: BRCA1 mutation detected - RNA: High PARP1 expression in tumor cells

Insight: PARP inhibitor (olaparib) likely effective

Mechanism: BRCA1 mutation + PARP inhibition = synthetic lethality

2. Immunotherapy Selection

Patient Profile: - DNA: High tumor mutational burden (TMB) - RNA: CD8+ T cell infiltration + PD-L1 expression

Insight: Strong immunotherapy candidate

Rationale: Neoantigens + active immune response

3. Drug Resistance Prediction

Patient Profile: - DNA: EGFR L858R mutation - RNA: MET amplification (expression 10x normal)

Insight: EGFR inhibitor resistance likely via MET bypass

Strategy: Combine EGFR + MET inhibitors

Technical Architecture

Data Integration Pipeline

graph LR
    A[BAM Files] -->|DeepSomatic| B[VCF Variants]
    C[scRNA Matrix] -->|Cell2Sentence| D[Cell Types]
    B --> E[Multi-Omics Engine]
    D --> E
    E --> F[Integrated Insights]

Analysis Layers

  1. Variant-Expression Correlation
  2. Map mutations to gene expression changes
  3. Identify pathway dysregulation

  4. Predict functional impacts

  5. Tumor Microenvironment

  6. Correlate mutations with immune infiltration
  7. Assess immunogenic potential

  8. Predict treatment response

  9. Pathway Analysis

  10. Integrate KEGG/Reactome pathways
  11. Find convergent alterations
  12. Identify therapeutic targets

Available Features

Currently Live ✅

  • [x] Pathway-Level Multi-Omics Correlation 🆕 (v1.1.0)
  • [x] KEGG and Reactome Pathway Enrichment 🆕 (v1.1.0)
  • [x] Automated Druggable Target Identification 🆕 (v1.1.0)
  • [x] DNA variant analysis with deep learning
  • [x] Single-cell RNA-seq analysis with Cell2Sentence
  • [x] Multi-omics integration dashboard
  • [x] Population genetics enrichment (71K+ genomes)
  • [x] Protein structure predictions (200M+ proteins)
  • [x] Drug target discovery (1.6M+ compounds from ChEMBL)
  • [x] Clinical evidence and therapeutic matching
  • [x] Medical imaging integration (TCIA/GDC)
  • [x] Unified visualization dashboard with sticky navigation
  • [x] CSV export with 50+ enriched columns
  • [x] Professional PDF reports with Gamma Generate

Coming Features

Q2 2026

  • [ ] Spatial transcriptomics support
  • [ ] Proteomics integration (mass spec)
  • [ ] Advanced drug response prediction models
  • [ ] Enhanced clinical trial matching

Q3 2026

  • [ ] Longitudinal analysis (pre/post treatment)
  • [ ] Multi-sample comparison and batch analysis
  • [ ] Custom pathway analysis
  • [ ] Clinical-grade PDF reports

Example Workflow

Step 1: DNA Analysis

Upload tumor BAM → Get somatic variants:

TP53 chr17:7577538 G>A (p.R273H)
KRAS chr12:25398284 C>T (p.G12D)

Step 2: RNA Analysis

Upload scRNA-seq → Get cell types:

CD8+ T Cells: 35%
Cancer Cells: 40%
Macrophages: 15%

Step 3: Multi-Omics Integration

Automated cross-analysis now available:

TP53 R273H Mutation: - Found in 40% of cancer cells (scRNA) - Associated with MDM2 overexpression (2.5x) - Recommendation: Consider MDM2 inhibitor (Idasanutlin)

KRAS G12D Mutation: - Found in cancer cell cluster 2 - MAPK pathway genes upregulated - Recommendation: MEK inhibitor (Trametinib)

Immune Landscape: - High CD8+ T cell infiltration - PD-1/PD-L1 axis active - Recommendation: Immunotherapy eligible

API Preview

Future programmatic access:

from cancerscope import MultiOmicsAnalysis

# Initialize analysis
analysis = MultiOmicsAnalysis()

# Add DNA data
analysis.add_dna(
    tumor_bam="tumor.bam",
    normal_bam="normal.bam",
    model="WGS"
)

# Add RNA data
analysis.add_rna(
    expression_matrix="cells.csv",
    model="c2s-gemma-27b"
)

# Run integrated analysis
results = analysis.run()

# Get treatment recommendations
treatments = results.get_recommendations()
print(treatments)

Data Requirements

For DNA Analysis

  • Tumor BAM file (aligned reads)
  • Normal BAM file (optional but recommended)
  • Reference genome: GRCh38

For RNA Analysis

  • Expression matrix (genes × cells)
  • Cell metadata (optional)
  • Gene annotations (HGNC symbols)

For Integration

  • Both DNA and RNA from same sample
  • Matched tissue/timepoint
  • Quality metrics passing thresholds

Clinical Applications

Biomarker Discovery

Identify novel biomarkers: - Mutations + expression signatures - Predictive of treatment response - Prognostic indicators

Treatment Selection

Evidence-based therapy choice: - Match mutations to targeted drugs - Assess immunotherapy potential - Avoid resistance mechanisms

Clinical Trials

Find eligible trials: - Mutation-based inclusion criteria - Expression-based stratification - Combination therapy studies

Resources

Feedback

Help us prioritize features! Contact us with: - Desired integrations - Use case requirements - Feature requests

Status: Available Now
Features: DNA somatic variant analysis, single-cell RNA-seq, multi-omics integration dashboard