PandaOmics

PandaOmics is an AI-powered omics data analysis platform developed by Insilico Medicine, designed to support drug discovery teams, academic researchers, and biopharmaceutical organizations in target identification, indication prioritization, biomarker discovery, and drug repurposing. By combining multi-modal AI models with a comprehensive knowledge base drawn from omics data, publications, clinical trials, grants, and patents, PandaOmics delivers data-driven, explainable insights to accelerate early-stage research decisions.

The platform is structured around three core pillars: Target ID, Data Analysis, and Knowledge Base. Together, these capabilities enable researchers to move from raw omics data through pathway-level understanding to actionable gene-disease association reports, all within a single environment. An academic subscription is available at $199 per month, and an on-premise deployment option is also offered for organizations requiring local infrastructure.

Target Identification Capabilities

• Actionable Target Identification: Discover disease-relevant targets using omics data, publications, clinical trials, and additional evidence sources. The AI prioritizes genes through comprehensive multi-modal models, allowing users to tune rankings based on preferences for novelty, safety, and the availability of existing drugs.
• In-house Data Integration: Upload proprietary omics datasets to refine and personalize target rankings alongside publicly available data.
• Target Evaluation: Systematically assess background evidence for shortlisted targets by analyzing both molecular and text-based evidence connecting a target to a specific disease.
• Target Attention Prediction: Evaluate historical trends in scientific interest for a given gene and forecast how attention is expected to grow in upcoming years, helping teams prioritize targets with emerging momentum.

Data Analysis Features

• Omics Data Analysis: Access normalized and annotated omics data collections, removing the burden of pre-processing and data hosting. Run target identification, indication prioritization, and compound prioritization workflows, and analyze differential expression to identify significantly over- or under-expressed genes across sample groups.
• Pathway Analysis: Identify molecular processes affected by disease using the proprietary iPanda algorithm. Pathway analysis provides mechanistic insights into disease biology and supports both publicly available and custom-uploaded omics datasets.
• Biological Knowledge Graph: Leverage large language models to surface insights from scientific literature about genes and diseases. Relevant publications are visualized as a Knowledge Graph for rapid literature navigation. The integrated ChatPandaGPT assistant allows users to ask questions and receive answers quickly, saving time on manual literature reviews.
• Gene-Disease Association Reports: Automatically generate comprehensive reports summarizing all relevant information about a gene-disease association, including supporting omics evidence, mechanism of action, competitive landscape analysis, and suggested path to market.

Knowledge Base Scale

• 1.3 million disease-specific omics samples covering a full spectrum of data generated by the scientific community
• 15,000 clinical-stage compounds and biologics, spanning Phase 1 through FDA-approved drugs
• 3.2 million life sciences research grants
• 5.5 million patents covering the life sciences industry
• 998,000 clinical trial records providing additional context on trial design and outcomes
• 47 million published biomedical research results

Validated Use Cases and Research Applications

• Identification of TNIK as an anti-fibrotic target using PandaOmics AI, validated in preclinical and clinical models
• Discovery and prioritization of novel aging-associated targets for drug discovery using hallmarks-of-aging-based dual-purpose disease analysis
• Biomarker identification, therapeutic repurposing, and new drug discovery for conditions such as diabetes, as demonstrated by the Indiana Biosciences Research Institute
• Rapid analysis and interpretation of Microarray and RNA-seq data for target identification workflows

Collaborations and Partnerships

• Collaboration with invoX Pharma to enhance R&D using AI (2024)
• Partnership with SRW to advance longevity science through AI (2023)
• Joint research with the University of Copenhagen and University of Chicago on premature aging diseases and cancer (2022)
• Partnership with Taisho on AI-powered senolytic drug discovery (2020)
• Collaboration with Boehringer Ingelheim on AI systems for target discovery (2020)
• Collaboration with Pfizer on novel data and AI systems for target discovery (2020)

PandaOmics is published in the Journal of Chemical Information and Modeling (February 2024) as an AI-driven platform for therapeutic target and biomarker discovery. In addition to its cloud-based offering, PandaOmics is available as an on-premise universal system AI device (BOX), enabling organizations to deploy the platform within their own infrastructure to support target identification, biomarker discovery, and indication prioritization programs.