PapersFlow for Chemistry, Drug Discovery & Structural Biology
A unified research platform with direct access to ChEMBL (2.5M+ compounds, 13M+ bioactivity records), PubChem compound properties, ZINC (230M+ purchasable compounds), PDB protein structures, and UniProt (250M+ protein entries). Search databases, retrieve bioactivity and structural data, and visualize molecules as interactive 2D SMILES renderings — all within a single AI-powered research session alongside the scientific literature.
Query ChEMBL bioactivity, PubChem properties, ZINC purchasable compounds, PDB protein structures, and UniProt protein annotations — with inline 2D SMILES molecule visualization — all from one AI research assistant.
Chemistry and drug discovery research requires constant context-switching between siloed databases. You search ChEMBL for bioactivity data in one tab, check PubChem for compound properties in another, browse ZINC for purchasable analogs, look up PDB for protein-ligand structures, and cross-reference UniProt for protein annotations — all while tracking the primary literature in yet another window. Each database has its own query syntax, its own output format, and no connection to the papers you are reading. PapersFlow eliminates this fragmentation by integrating all five databases into a single conversational interface alongside 474M+ scientific papers, so you can move from literature finding to compound lookup to structural analysis without ever switching tools.
What You Can Do
- ChEMBL Drug Discovery Database
- PubChem Compound Properties
- ZINC Database (230M+ Compounds)
- PDB Protein Structures
Tools
Compare
Frequently Asked Questions
- What is the difference between ChEMBL, PubChem, and ZINC?
- Each database serves a different purpose. ChEMBL contains curated bioactivity data — IC50, EC50, Ki, Kd values from published assays — making it the go-to resource for understanding how compounds interact with biological targets. PubChem provides physicochemical compound properties such as molecular weight, LogP, TPSA, and hydrogen bond counts, useful for assessing drug-likeness and ADMET characteristics. ZINC catalogs 230M+ commercially available compounds with vendor and purchasability information, making it essential for finding buyable analogs of hit compounds. PapersFlow lets you query all three from the same research session.
- Can I search by SMILES notation?
- Yes. The ZINC database integration supports SMILES similarity search using Tanimoto distance, so you can paste a SMILES string and find structurally similar purchasable compounds. Additionally, all compound results from ChEMBL, PubChem, ZINC, and PDB ligand queries that include SMILES data are automatically rendered as interactive 2D molecule diagrams with element-specific coloring, so you can visually inspect structures without leaving the conversation.
- Does it include protein structure data?
- Yes. PapersFlow integrates both PDB and UniProt. The PDB integration lets you search for protein structures by text or sequence and retrieve metadata such as resolution, experimental method, atom count, and bound ligand details including ligand SMILES, InChI, and molecular formula. UniProt provides protein sequences, functional domain annotations, post-translational modifications, and cross-database ID mapping. Note that PapersFlow returns structural metadata and ligand information — it is not a 3D molecular viewer, but it connects structural data directly to your literature research.
- Does it support ACS citation format?
- Yes. Chemistry outputs default to ACS citation style with superscript numbered references. You can export formatted reference lists, .bib files, and complete manuscripts styled for JACS, ACS Nano, Chemistry of Materials, and other ACS publications.