Insilico Medicine (Pharma.AI)

Upload patient RNA-seq data to PandaOmics.

Run target identification against 'Normal vs Tumor' differential expression.

Filter for 'High Druggability' and 'Low Tissue Specificity' for safety.

Validate top 3 targets via the integrated Knowledge Graph.

Define target protein pocket in Chemistry42.

Select generative models optimized for scaffold hopping.

Set reward functions for blood-brain barrier permeability.

Synthesize top 5 generated compounds for in-vitro testing.

Input drug candidate and trial protocol into InClinico.

Analyze historical trial performance of similar mechanisms of action.

Review 'Probability of Success' score and identified risk factors.

Adjust trial design (e.g., patient selection) based on AI recommendations.

Cross-reference rare disease pathway maps in PandaOmics.

Identify overlapping nodes with established therapeutic targets.

Score targets based on existing safety data.

Initiate virtual screening for approved drugs against these targets.

Utilize PreciousGPT to analyze longevity datasets.

Rank targets based on impact on aging hallmarks.

Generate molecules targeting senolytic pathways.

Analyze output for systemic safety profiles.

Input lead structure into Chemistry42 optimization module.

Run structure-activity relationship (SAR) analysis.

Generate analogs with modifications focused on blocking metabolic sites.

Rank analogs by predicted clearance rates and potency.

Analyze clinical trial responder vs non-responder omics data.

Identify transcriptomic signatures associated with response.

Develop a predictive biomarker panel in PandaOmics.

Implement panel in Phase III trial recruitment.