Page 1
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
InfoChem Product Presentation
ICIC - International Conference for the Information Community
Heidelberg, Germany, October 17, 2016
Dr. Valentina Eigner-Pitto
1 / 14
Page 2
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
InfoChem at a Glance
People
• 21 full time employees
(Munich office)
• 1 consultants in UK
• 60 freelance abstractors
(residing offshore)
Company
• specialized in chemoinformatics
• founded in 1989
• based in Munich, Germany
• owned by Springer Nature
Business Areas
• Software products
• Projects
• Text/Data mining
• Database building
Customers
• Pharmaceutical industry
• Chemical industry
• Scientific publishers
• Academia
• IP Professionals
Company overview 2 / 14
Page 3
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Data Mining Projects
Services
• Project development
• Consulting
• Database building
• Chemical entity recognition
• Name to structure conversion
• Image to structure conversion
• ChemDraw CDX files work-up
Software
• ICFSE, ICCARTRIDGE, ICCHEMDESK
• ICMAP, CLASSIFY, ICNameRXN
• ICSYNTH, ICFRP, ICEDIT, ICTOOLS
• Markush…
Business Areas
Company overview
Content
• SPRESIweb, SPRESImobile
• Patents / Structures
• Chemisches Zentralblatt
Structural Database
3 / 14
Page 4
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Services
Content Data Mining Projects
• Project development
• Consulting
• Database building
• SPRESIweb, SPRESImobile
• Patents / Structures
• Chemisches Zentralblatt
Structural Database
• Chemical entity recognition
• Name to structure conversion
• Image to structure conversion
• ChemDraw CDX files work-up
Software
• ICFSE, ICCARTRIDGE, ICCHEMDESK
• ICMAP, CLASSIFY, ICNameRXN
• ICSYNTH, ICFRP, ICEDIT, ICTOOLS
• Markush…
Business Areas
Company overview
MARKUSH
4 / 14
Page 5
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
InfoChem Fast Search Engine: ICFSE
• High performance chemistry search engine
• Retrieval of data from millions of records in seconds
• Easy integration in any desktop or web application with
multiple platform support
• Supports typical query features and search types for
structures and reactions
o atom query: any atom, heteroatom, list/not list...
o bond query: bond type, topology...
o reacting center query: change, make/break...
o search types: exact, substructure, tautomer, isomer,
flexmatch, similarity, all-in-one...
Search engine for Markush 5 / 14
Page 6
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Deployment in New STN
Lecture ICIC 2015 Fiz-
Karlsruhe, Thomson Reuters
The Driving Force
First milestone
• Definition of specific data
format for Markush structures
• First prototype for storage
and retrieval
Search engine for Markush
2008 - 2010 2015 2012 …
MARKUSH
Cooperation goal:
• Integrate in New STN
o generic searches
o Markush searches of
DWPIM
Future developments:
• Search functionalties
o Nested R-groups
o Markush query
• Performance
• …
6 / 14
Page 7
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Structure Searchable Representation of Markush
• Generation of a specific structure representation to
enable structure searches in the ICFSE index
o normalisation of text to atom properties
o evaluation and enumeration of variations (s, p, f, h) s-variation: R1 = methyl or ethyl
p-variation: R3 = amino
f-variation: n= 1 – 3
h-variation: R2 = alkyl [Size(3-9);TB(>0)]
MARKUSH Search engine for Markush 7 / 14
Page 8
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
A Novel Concept for the Search and Retrieval of the Derwent Markush
Resource Database Andreas Barth,*,† Thomas Stengel,† Edwin Litterst,† Hans Kraut,‡ Henry Matuszczyk,‡ Franz Ailer,‡ and Steve Hajkowski§ †FIZ Karlsruhe − Leibniz Institute for Information Infrastructure, D-76344 Eggenstein-Leopoldshafen, Germany
‡InfoChem GmbH, D-81241 Munich, Germany
§Thomson Reuters, London EC1N 8JS, United Kingdom
Product Match Levels
STN match level concept:
• ATOM retrieves only specific nodes (standard for ring nodes):
o Specific atoms in the query match only specific atoms in the file
o Generic groups in the query match only specific atoms/groups in
the file
• CLASS retrieves both specific and generic nodes (standard for
chain nodes):
o Specific atoms/generic groups in the query match to specific or
generic atoms/groups in the file
• ANY retrieves specific and generic nodes plus the R (XX) node
MARKUSH
Mechanism to control the searching level in Markush structures (Marpat, Questel)
Search engine for Markush 8 / 14
Page 9
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Hit Visualization
MARKUSH
S
O
Query:
Search engine for Markush 9 / 14
Page 10
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Hit Visualization: Highlighting
MARKUSH
S
O
Search engine for Markush
Query:
10 / 14
Page 11
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Hit Visualization: Assembled Hit
MARKUSH
S
O
Search engine for Markush
Query:
11 / 14
Page 12
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
Support of Superatoms
MARKUSH
Acyclic Cyclic
Ak (Chain) Cy (Ring)
Cb (Carbcycle)
Hy (Heterocycle)
CHK (Alkyl, Alkylene)
CHE (Alkenyl, Alkenylene)
CHY (Alkynyl, Alkynyline)
CYC (Cycloaliphatic)
HEA (Monocyclic heteroaryl)
HEF (Fused heterocyclic)
HET (Monocyclic nonaromatic)
ARY (Aryl)
Search engine for Markush 12 / 14
Page 13
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
New @ InfoChem
Reaction prediction tools
• Further development and optimisation of the algorithm
• Design of a collaborative platform enabling user team work
Other news from InfoChem
Data mining
• ICANNOTATOR extended to further languages:
o French
o Chinese, Japanese, Korean (in cooperation with
NextMove Software)
13 / 14
Page 14
InfoChem Copyright © 2016 Dr. Valentina Eigner Pitto Product Presentation ICIC 2016, Heidelberg, Germany, October 17
InfoChem GmbH: www.infochem.de, www.spresi.com, [email protected]
Visit us at the InfoChem booth!
14 / 14