CORTEX+SYNAPSE

Toxicogenomics & Safety

Evaluate compound safety by combining your experimental toxicogenomics data with real-world pharmacovigilance signals. Synapse consolidates FAERS reports, clinical trial adverse events, and class-wide liability patterns across related drugs. Cortex analyzes your treated-vs-control expression data and maps changes onto known toxicity markers. The integration reveals whether your compound's signals match known class liabilities or represent novel concerns.

What You Get

Deliverables

Integrated risk assessment

Combines your experimental expression changes with FAERS pharmacovigilance signals and clinical trial safety data into a single organ-level risk view. Each risk call is supported by evidence from multiple independent sources: your own data, real-world adverse event reports, and published clinical observations. For the GLP1R class, this identified 2 medium-risk organ systems (CNS, GI) with class-wide liability patterns across 9 of 17 marketed drugs.

SystemicRespiratoryRenalMusculoskeletalMetabolicImmuneHepaticHematologicDermatologicCardiacGICNSLowLowLowLowLowLowLowLowLowLowMediumMedium

Organ-system safety map

Your compound's expression signature overlaid onto known organ-specific toxicity markers across 12 organ systems. Expression changes in toxicity-associated genes are flagged and compared against the class baseline. The pharmacogenomic drug screen data (218 drugs, 17,546 genes) provides the reference: fenofibrate alone showed 2,617 differentially expressed genes, many mapping to known hepatotoxicity markers.

PancreatitisConstipationDiarrhoeaVomitingNausea0k10k20k30k40k50k60k8k13k25k26k52k

Class liability detection

Systematic comparison of your compound's signals against patterns observed across related marketed drugs. The analysis distinguishes expected pharmacology (e.g., GI effects from GLP-1 agonism) from potentially novel liabilities. A fibrate neurotoxicity signal, TRAIL/death receptor signaling activation, was identified across ciprofibrate (z=6.28) and fenofibrate (z=6.11), demonstrating how cross-compound comparison reveals hidden safety signals.

DermatologicGICNS8/179/179/17
DECISION ENABLED

Prioritize compounds with cleaner safety profiles and flag liabilities before they become expensive surprises in clinical development.

Sample Output

GLP1R class safety profile + drug screen findings

GLP1R: Organ System Risk Grid12 organs assessed
SystemicRespiratoryRenalMusculoskeletalMetabolicImmuneHepaticHematologicDermatologicCardiacGICNSLowLowLowLowLowLowLowLowLowLowMedium (Class liability)Medium (Class liability)
FAERS Class Liabilities: Top Events by Report Count9/17 drugs affected
InsomniaTremorAbdominal painPancreatitisDizzinessHeadacheConstipationDiarrhoeaVomitingNausea010k20k30k40k50k60k3,1293,3978,1368,17710,66711,38513,31424,55825,80551,812
GI system CNS system
Pharmacogenomic Drug Screen: Top Compounds218 drugs screened
DrugDEGsTop Pathway
Fenofibrate2,617Lipid metabolism
Nilotinib2,114XBP1 (z=16.4)
Mitoxantrone1,343DNA damage response

Key finding: Only 50% directional agreement between analytical methods, demonstrating why multi-method consensus analysis is essential.

CROSS-TARGET COMPARISON: IL4R / DUPILUMAB

Multi-Target FAERS Comparison: Top 5 EventsGLP1R class vs IL4R
FAERS reports0k10k20k30k40k50k60kTop event2nd3rd4th5thGLP1R classIL4R/Dupilumab

IL4R/Dupilumab generates massive FAERS volume from a single marketed drug (pruritus 48,441 reports). GLP1R class signal is distributed across 17 drugs. Different denominator structures require careful interpretation.

IL4R/Dupilumab: FAERS Event Detail7 class liabilities
CNSDermatologicGIImmuneMusculoskeletalRespiratorySystemic
EventReports
Pruritus48,441
Dermatitis Atopic37,675
Rash33,431
Dry Skin21,662
Arthralgia16,570
Dyspnoea11,604
Injection Site Erythema9,484
Cough9,331
Eye Pruritus8,919
Cyclosporine Fibrosis Signal: AD Analysis (GSE157194)safety flag
log2FCCOL1A1COL1A2COL3A100.511.522.5+2.28+1.89+2.29

Cyclosporine upregulates collagen genes (COL1A1 +2.28, COL1A2 +1.89, COL3A1 +2.29), a fibrosis safety signal detected from transcriptomic data that complements traditional pharmacovigilance. Cyclosporine had 22 total safety flags vs only 5 for dupilumab in the same AD cohort.

How It Works

Methodology

STEP 1

Synapse pulls pharmacovigilance signals

Consolidated FAERS reports and clinical trial adverse events for the entire drug class. For GLP1R, this covers 17 drugs with event-level data across 12 organ systems, identifying class-wide patterns from real-world post-market surveillance.

STEP 2

Cortex runs differential expression

DESeq2 analysis on your treated-vs-control samples identifies genes with significant expression changes. Multi-method consensus scoring (across analytical approaches) addresses the finding that only 50% of results show directional agreement between individual methods.

STEP 3

Expression mapped to toxicity markers

Differentially expressed genes are mapped onto curated organ-specific toxicity marker panels. Changes in known hepatotoxicity, cardiotoxicity, or neurotoxicity genes are flagged and ranked by effect size and statistical confidence.

STEP 4

Class liability comparison

Your compound's safety signal profile is compared against all related marketed drugs in the class. Shared patterns (like GI disturbance across GLP-1 agonists) are labeled as expected class effects. Novel signals absent from the class profile are flagged for further investigation.

STEP 5

Integrated risk assessment

Experimental and real-world evidence are combined into a single organ-level risk scorecard. Each risk call is annotated with the contributing evidence sources, enabling transparent safety decisions grounded in both your data and post-market experience.

Who This Is For

Target personas

Safety pharmacologist

Interpret compound-specific expression changes in the context of known class liabilities and real-world pharmacovigilance data.

Toxicology lead

Identify organ-specific risk signals early to guide preclinical study design and regulatory strategy.

Regulatory affairs

Build safety narratives grounded in integrated evidence from experimental data and post-market surveillance.