Plex

Plex is an AI-powered scientific intelligence platform built for pharmaceutical and biotech drug discovery teams. Rather than relying on large language model text summarisation, Plex is grounded in a proprietary biomedical knowledge graph — the Focal Graph — containing billions of structured data points spanning genomics, proteomics, compound activity, clinical trials, and published literature. The result is reproducible, evidence-backed answers that are directly traceable to specific scientific databases and records.

Trusted by more than 40 pharmaceutical and biotech discovery teams, Plex is designed to support decisions across the full R&D lifecycle — from early-stage target identification and biomarker discovery through to portfolio-level governance. It serves scientists seeking to accelerate hypothesis generation, developers looking to integrate knowledge graph capabilities into computational pipelines, and executives who need structured, confidence-scored evaluations before committing capital.

Core Platform Capabilities

• Knowledge Graph (Focal Graph): Billions of data points from genomics, proteomics, clinical trials, and literature are structured and queryable in real time, enabling direct interrogation of biological and chemical datasets rather than text search.
• Evidence Scoring: Every answer is backed by quantified evidence assessed across seven evaluation dimensions, with full source traceability so outputs can be audited and reproduced.
• Proprietary Data Integration: Internal experimental data can be incorporated into Plex under strict confidentiality, with data never leaving the user's environment.
• Chemical Intelligence Tooling: Built-in structure-aware compound analysis tools allow users to query and interpret chemical datasets alongside biological evidence.
• Multi-Omics Integration: Genomic, proteomic, and Perturb-seq signatures are integrated and ranked by evidence strength to surface the most relevant targets and biomarkers.

How Plex Works

1. Ask a question: Users pose any scientific question — from target identification to biomarker discovery. Plex understands both biological and chemical context, supporting queries such as evaluating a specific target for a disease indication or identifying resistance mechanisms in a given cancer type.
2. Plex searches the knowledge graph: Billions of data points spanning genomics, proteomics, compound activity, and clinical data are integrated and scored in real time across the Focal Graph.
3. Receive structured, actionable answers: Results are delivered with confidence scores, source citations, and chemical intelligence tooling, ready for use in R&D decision-making and governance processes.

Plex vs. Literature-Based LLMs

• Literature-based LLMs are trained on text and search published documents rather than structured scientific databases, generating plausible-sounding but unverifiable answers with no access to genomic, proteomic, or chemical datasets and no reproducibility across queries.
• Plex is built on billions of structured data points from biomedical databases, queries genomic, proteomic, clinical, and chemical datasets directly, and produces reproducible, evidence-scored outputs where every answer is traceable to specific data sources and records.

Use Cases

• Target Identification: Discover novel therapeutic targets supported by multi-omics evidence integration.
• Biomarker Discovery: Identify predictive and prognostic biomarkers across patient populations.
• Target Evaluation: Score and compare targets across seven evidence dimensions before committing capital resources.
• Scientist Workflows: Accelerate hypothesis generation with structured, evidence-backed analysis.
• Developer Integration: Incorporate Plex's knowledge graph into existing computational pipelines via API.
• Executive Decision-Making: Enable faster, more confident portfolio decisions through structured evaluation outputs.

Decision-Ready Scientific Deliverables

• Target Evaluation Report: A structured assessment across seven dimensions — biology, genetics, druggability, safety, clinical signal, competitive landscape, and proprietary data concordance — with multi-omics evidence scoring and PDF export.
• Biomarker Panel: Prioritised candidate biomarkers with predictive power estimates, validation status, and recommended assay strategies for clinical translation, covering predictive, prognostic, and companion diagnostic applications.
• Competitive Intelligence: Real-time landscape analysis of competing programmes, mechanism-of-action overlap, clinical trial status, patent analysis, and differentiation opportunities.

Plex is designed to integrate seamlessly with internal experimental data while maintaining strict data confidentiality, making it suitable for organisations that require both external biomedical breadth and secure handling of proprietary research assets. The platform delivers structured intelligence in formats directly applicable to governance and portfolio decision-making processes.