Transcriptomics
Gene expression profiles across tissues, diseases, and compound treatments
For Biologists & Chemists
Ask a question.
See the evidence.
Plex searches billions of experimental data points across multi-omics, chemical biology, and literature — powered by Focal Graphs and LLMs. Every result traces back to real data.
The Problem
You shouldn’t need 12 tabs open to test one hypothesis.
Today, testing a single biological hypothesis means manually searching fragmented databases, reconciling conflicting nomenclature, and assembling evidence from sources that were never designed to talk to each other.
Hours of manual synthesis
Searching one target across transcriptomics, proteomics, chemical bioactivity, and literature takes days of manual curation.
No cross-source linking
A gene expression signal in one database has no automatic connection to a compound screen or clinical finding in another.
Evidence is buried
The critical data point that changes your hypothesis might exist in a 2019 proteomics dataset you never knew about.
Core Technology
The Focal Graph: billions of data points, one unified structure.
Inspired by the centrality algorithms Google uses to rank the web, the Focal Graph applies the same principle to biomedical research — identifying the strongest connections across diverse, disparate experimental datasets.
- Centrality-ranked results — not keyword matches, but statistically significant convergence of evidence across sources
- Minimally processed data — works on raw experimental outputs, not curated summaries that lose signal
- Cross-source linking — a transcriptomic signal automatically connects to chemical bioactivity, proteomic evidence, and clinical findings
No black boxes. Every hypothesis is linked to the experiments that support it.
Integrated Data
Many data dimensions. One search.
Plex covers billions of data points across every major experimental modality — from raw omics to chemical structures. Plus, securely integrate your own proprietary data.
Proteomics
Protein-level measurements, interactions, and post-translational modifications
Metabolomics
Metabolite profiles and pathway perturbation signatures
Genomics
Genetic associations, variants, and functional genomics data
Chemical Biology
Small molecule bioactivities, compound structures, and SAR data
Scientific Literature
Published findings, clinical studies, and mechanistic evidence
+
Your Proprietary Data
Securely and privately integrate your organization’s internal experimental data. Analyzed through the same Focal Graph framework, your findings are weighted alongside the broader evidence landscape.
How It Works
From question to traceable evidence in seconds.
01
Ask a question
Use natural language to describe your hypothesis, target, or biological question. Plex understands the science behind the words.
"Which kinases show convergent evidence across transcriptomic and chemical screens in AML models?"
02
Focal Graph searches
Centrality algorithms traverse billions of data points across all integrated sources simultaneously, ranking results by evidence strength — not keyword frequency.
03
LLMs synthesize
LLM’s mine and interpret graph results provided by Plex, delivering detailed answers in the language scientists use — with every claim linked to supporting experimental data.
04
Trace to evidence
Every result links back to the specific experimental datasets in Plex that support it. Follow the chain from insight to raw data, evaluate the evidence yourself, and design targeted follow-up.
Let the AI run the research campaign.
AutoPlex is an agent-based AI system that autonomously plans and executes complex drug discovery workflows. It runs hundreds of Focal Graph queries, interprets results, refines strategy, and delivers a complete research report — with full documentation of every step taken and every data point uncovered.
Autonomous planning
Describe a research objective in natural language. AutoPlex decomposes it into a multi-step query plan, executes it, and iterates based on what it finds.
Large-scale campaigns
Run research programs that would take a team weeks — systematic target sweeps, indication expansions, or safety profiling across entire compound libraries.
Full audit trail
Every AutoPlex result includes the complete strategy used, the queries executed, and explicit links to all supporting experimental evidence. Nothing is a black box.
Applications
What scientists ask Plex.
Whether you’re a biologist testing a mechanistic hypothesis or a chemist evaluating a compound series, Plex answers questions that previously required weeks of manual work.
Target Identification & Validation
Find novel targets with convergent multi-omics evidence. Validate biological hypotheses against billions of data points before committing to wet lab experiments.
"Show me targets with convergent transcriptomic and proteomic evidence in NASH liver tissue"
Disease Indication Discovery
Uncover new disease indications for existing targets or compounds. Plex connects drug-disease relationships that span datasets no single researcher could traverse.
"What disease indications show unexpected compound bioactivity overlap with our lead series?"
Biomarker Discovery
Identify patient selection biomarkers and pharmacodynamic markers by searching across genomic associations, expression profiles, and clinical data simultaneously.
"Which biomarkers discriminate responders from non-responders in JAK inhibitor trials across RA and UC?"
Safety Assessment
Surface off-target risks and safety signals early by cross-referencing chemical bioactivity with expression profiles, known toxicology, and clinical adverse events.
"What off-target activity does this chemical series show across hepatocyte and cardiomyocyte expression panels?"
Transparency
Every insight links back to the data.
Unlike black-box AI systems that generate plausible-sounding answers, Plex’s uniquely transparent approach means every finding is explicitly linked to supporting experimental data points.
- Follow the evidence chain from high-level insight down to individual measurements in source datasets
- Evaluate the quality and relevance of supporting evidence yourself
- Design targeted follow-up experiments based on exactly what the data shows — and what it doesn’t
- Share findings with colleagues with full provenance, not just AI-generated text
FAQ
Questions scientists ask about Plex.
What is the Plex Focal Graph?
The Focal Graph is Plex's proprietary knowledge graph that uses centrality algorithms — inspired by Google's PageRank — to identify the strongest connections across billions of biomedical data points. It distills vast, noisy datasets into concise, transparent, data-driven hypotheses with every finding linked to supporting experimental evidence.
What data sources does Plex search?
Plex searches across transcriptomics, proteomics, metabolomics, genomics, chemical structures, small molecule bioactivities, gene expression profiles, clinical studies, and scientific literature. Organizations can also securely integrate their own proprietary experimental data.
What is AutoPlex?
AutoPlex is Plex's agent-based AI system that autonomously plans and executes complex drug discovery research programs. It can run large-scale research campaigns, interpret results, and provide full details of the strategy used and the experimental validation data uncovered.
How does Plex use large language models?
Plex combines the Focal Graph with LLMs to create an AI that understands chemical structures and omics data. Scientists ask questions in natural language, and the system returns data-driven answers with explicit links to the supporting experimental evidence — not just text summaries, but traceable results from real datasets.
Can I integrate my own proprietary data into Plex?
Yes. Plex allows organizations to securely and privately integrate their own proprietary experimental data alongside billions of public data points. Your internal data is analyzed through the same Focal Graph framework, enabling discoveries that bridge internal findings with the broader biomedical evidence landscape.
Try Plex
See your data through the Focal Graph.
Request a live demo with your own research question. We’ll run it through Plex in real time and show you what the Focal Graph uncovers.
BillionsData points
40+Companies served
6Omics modalities