How it works
Three steps, one list you can hand over
- 1
Give the structure
Draw it in the editor, or paste a name, CAS number, SMILES or InChI. Every search type works from either input.
- 2
We search the public records
Exact, substructure and similarity matching against PubChem compounds, then a wider scan of related compounds so patents on analogs, intermediates and generic examples are collected too.
- 3
Read, filter, export
Patents open as PCT families with titles, abstracts, assignees and priority dates. Filter and sort, then export to CSV, PDF or Word.
What you get
Built for an IP desk, not a shopping cart
Exact match
Same atoms, bonds, stereochemistry and isotopes — the quick novelty check before a full search.
Substructure
Everything that contains your drawn fragment, up to 200 compounds per run.
Similarity
Close analogs ranked by 2D Tanimoto, with the threshold you set (70–99%).
Markush-aware scan
Patents from related compounds are merged automatically: analogs, intermediates and smaller examples of generic formulas (R = alkyl, cycloalkyl, heterocycle).
PCT family view
WO families lead the list; click a PCT number to open the whole family with titles and dates.
Filters & export
Filter by number and priority date, sort ascending or descending, export CSV in full, PDF or Word at 100 results per file with a range picker.
Data & limits
Public records only — and it says so on every page
Where the data comes from
PubChem compound, patent and literature records, and PubMed abstracts. Nothing is scraped from paid databases, and no account or key is needed.
What it cannot tell you
Titles, dates and families are for orienting you, not legal status — confirm in Espacenet, WIPO, USPTO or the Indian Patent Office registers. Patents that mention a molecule only by name, with no deposited structure, are outside a structure search; the Google Patents and Espacenet links on each row are the fastest free way to those.
