Chemical Patent Curation and Management

New Tools and Capabilities

Árpád Figyelmesi, Daniel Bonniot de Ruisselet

Motivation

Knowing the chemical space covered by competitors’ patents is essential for successful drug discovery.

This information can be useful for: ● Idea generation ● Lead candidates selection ● Drug design ● Patent claims construction

Challenges

● The existing databases are optimized for patent searching

● Manual processing and analysing is time consuming and needs special expertise

● Automatic processing quality is not good enough ● No tools for editing complex Markush structures

Computer-assisted data extraction

● English, Chinese and Japanese Name to Structure dramatically speeds up the extraction process

● Markush Editor helps to draw complex Markush structures

● Structure Checker and Markush Validation guarantees the high quality of extracted information

● Markush Representation, Search and Enumeration

Name to Structure

● Support for many nomenclatures (common, drug names, …) ● IUPAC names used for exemplified structures ● Essential to extract chemical information from patents

● English (2008, Marvin 5.1) ● Chinese (2013, Marvin 5.12) ● Japanese (2014, Marvin 6.3)

Why other languages?

Validation on patent data

Measuring overlap between English and Chinese patents Using different data sources and tools

Document Annotation

Document Annotation

● Upcoming API in 6.4 ● Display annotated patents and documents (PDF, XML) ● Integrated in ChemCurator

Markush Editor Functions

● Editing complex patent markush structures ● Hierarchical representation of fragments’ relationships ● Visualization of nesting view, preview ● Editing separately the individual fragments ● Integrated structure checker ● Available as a desktop application and as an

integratable component (6.3)

Markush Editor

R-group definitions

Tree view

Scaffold

Structure checker

Nesting view & Preview

ChemCurator Functions

● Compound and Markush editor component ● Annotated documents from pdf, html, xml. ● Drag and drop structures from the document ● Connection between the document and extracted data ● Markush validation against the examples ● Support for multi-display environment ● Available as a desktop application (6.4)

ChemCurator

Markush editor

Example structures

Annotated document

Project explorer

Selected structures

Structure checker

Workflow

IP experts ● Search ● Analyze

IP experts ● Extract ● Validate

Database ● Markushes ● Examples ● Documents

Drug discovery team ● IJC ● Plexus ● JChem for Office

● IP experts can represent the chemical space ● Chemical representation is comprehensible for Medicinal Chemists ● High quality project specific database ● New opportunities, less risk, faster communication

Non patent Curation

Mode optimized for extracting specific structures ● Exemplified structures in patents ● Scientific journal articles ● Internal company reports ● … Wizard to automatically detect relevant structures ● Exclude fragments, chemical elements, …

Non patent Curation

Future plans

Naming: ● Keep improving accuracy in all languages (en,cn,jp) ● Add requested languages Markush: ● Overlap analysis ● Non-hit visualization ● Markush generation wizard ● Claim generation wizard

Acknowledgment