Overview of Reaxys Data
Last updated on April 10, 2025
Reaxys is the most comprehensive resource of the known chemistry space. It contains data on reactions, substances and bioactivities excerpted from 119 million documents. The machine ready data representation allows for applying the Reaxys content directly in data driven workflows. The Reaxys data corpus contains:
- 69 million reactions with experimental conditions and literature references
- 160 million substances and associated properties
- 48 million bioactivity data points and 43,000 biological targets
- 16,000 journals and 119 million documents including 45 million patents
The Reaxys data, delivered through the Reaxys API and use case specific flat files, can be embedded into a user's specific data pipelines and workflows to enable more accurate search, discovery, and prediction.
Reayxs API
- The Reaxys API provides access to the full Reaxys data corpus
- The Reaxys data is updated twice a week
Reaction Flat File (RFF)
- The RFF comprises 23 million ML ready reactions most relevant for small compound synthesis analytics and predictions
- Updated quarterly
Substance Structure Flat File (SFF)
- The SFF comprises the entire set of Reaxys chemical structures (56 million) excerpted from patents and journals
- Updated weekly
Reaxys Medicinal Chemistry Flat File (RMX FF)
- The RMC FF comprises the entire Reaxys dataset of reported bioactivities with associated substances, biological targets, and assays
- Updated weekly
FAQs
- ML and other models for predictive retrosynthesis, synthetic feasibility, reaction conditions, and reaction optimization
- Graph database of reaction paths
- Protein-ligand binding QSAR
- Reference data for FEP calculations
- Polypharmacology analytics
- Chemical structure analytics
- Structure-based models
- Reaxys data can be used for internal purposes exclusively
- Reaxys data can support semantic search capabilities via semantic enrichment
- Reaxys data can populate data networks such as new ontologies or knowledge graphs
- Reaxys data can be used to create predictive models
The update frequency varies by file type. The reaction flat file (RFF) receives a quarterly update, the RMC and structure flat files receive weekly updates. Data retrieved by the user via the Reaxys API is updated twice a week.
The Data for Research and Discovery Support team is available to assist with general onboarding, troubleshooting, documentation, and data interpretation questions. This team does not provide customer analysis of Elsevier data structures, recommendations on whether data is sufficient to meet project outcomes, or scoping of effort/estimates on data clean up and integration steps.
The Professional Services team is available for add-on services such as data integration with customer/third party data, custom workflows and dashboards, and custom reports. Please contact your Account Manager or Customer Consultant for more information about add-on services.
Did we answer your question?
Related answers
Recently viewed answers
Functionality disabled due to your cookie preferences