1. George Papadatos Mark Davies surechembl-help@ebi.ac.uk
SureChEMBL webinar
George Papadatos
Mark Davies

2. Outline SureChEMBL Interface demo Coverage and content Capabilities
Future plans
Interface demo

3. ChEMBL: Data for drug discovery
1. Scientific facts
3. Insight, tools and resources for translational drug discovery
>Thrombin
MAHVRGLQLPGCLALAALCSLVHSQHVFLAPQQARSLLQRVRRANTFLEEVRKGNLERECVEETCSYEEAFEALESSTATDVFWAKYTACETARTPRDKLAACLEGNCAEGLGTNYRGHVNITRSGIECQLWRSRYPHKPEINSTTHPGADLQENFCRNPDSSTTGPWCYTTDPTVRRQECSIPVCGQDQVTVAMTPRSEGSSVNLSPPLEQCVPDRGQQYQGRLAVTTHGLPCLAWASAQAKALSKHQDFNSAVQLVENFCRNPDGDEEGVWCYVAGKPGDFGYCDLNYCEEAVEEETGDGLDEDSDRAIEGRTATSEYQTFFNPRTFGSGEADCGLRPLFEKKSLEDKTERELLESYIDGRIVEGSDAEIGMSPWQVMLFRKSPQELLCGASLISDRWVLTAAHCLLYPPWDKNFTENDLLVRIGKHSRTRYERNIEKISMLEKIYIHPRYNWRENLDRDIALMKLKKPVAFSDYIHPVCLPDRETAASLLQAGYKGRVTGWGNLKETWTANVGKGQPSVLQVVNLPIVERPVCKDSTRIRITDNMFCAGYKPDEGKRGDACEGDSGGPFVMKSPFNNRWYQMGIVSWGEGCDRDGKYGFYTHVFRLKKWIQKVIDQFGE
Compound
Assay/Target
Bioactivity data
Ki = 4.5nM
APTT = 11 min.
2. Organization, integration, curation and standardization of pharmacology data

4. Why is searching chemical patents useful?
Infringement search to avoid areas of valid patent protection (freedom to operate)
Search for industrial profiles and research directions (competitive intelligence)
State-of-the-art/novelty/prior art search
Search for citations and key references
Most of the knowledge in chemical patents will never appear anywhere else
Average time lag between patent and journal: 3 years
Compounds, scaffolds, reactions
Biological targets, diseases, indications
Traditionally a closed and commercial field.
Average difference is 3 years
Comprehensive
IT has scaffolds and chemistry and compounds that will never appear or if it
Find novel claimed inhibitors of EGFR receptor published in 2014
Novel heterocyclic scaffolds / reaction schemes

5. SureChEMBL

6. SureChem becomes SureChEMBL
December 2013 EMBL-EBI acquired SureChem – a leading chemistry patent mining product from Digital Science, Macmillan Group
SureChem not aligned with core future academic business
Existing SureChem user base
Free (SureChemOpen)
Paying (SureChemPro + API)
EMBL-EBI supported existing licensees during transition
EMBL-EBI provides an ongoing, free and open resource to the entire community
Rebranded as SureChEMBL
Free users limited access
Full functionality for everyone

7. SureChEMBL data pipeline
SureChEMBL System WO EP
Applications& Granted US
Applications & granted JP
Abstracts
Patent
Offices
Chemistry Database
Entity Recognition
SureChem IP
1-[4-ethoxy-3-(6,7-dihydro-1-methyl-7-oxo-3-propyl-1H-pyrazolo[4,3-d]pyrimidin-5-yl)phenylsulfonyl]-4-methylpiperazine
Processed patents
(IFI Claims)
Name to Structure (five methods)
OCR
Database
Image to Structure
(one method)
Dynamic nature
Completely automated with no manual interaction – to be compared with chembl
Patent PDFs
(service)
Application Server
Users
API

8. SureChEMBL chemistry data coverage
Structures from text: 1976 onwards
Title, abstract, claims, description
IUPAC, trivial, drug names, etc.
SureChem Chemical Entity Recognition proprietary algorithm
ACD/Labs, ChemAxon, OpenEye, OPSIN, PerkinElmer name-to- structure conversion
Structures from images: 2007 onwards
CLiDE image-structure conversion
USPTO offers ‘Complex Work Units’ since 2001
CWU file types include MOL and CDX
CWUs processed as part of pipeline: 2007 onwards

9. SureChEMBL data content (11/03/2015)
16,261,347 unique compounds
13,274,991 chemically annotated patents
~80,000 novel compounds extracted from ~50,000 new patents monthly
2–7 days for a published patent to be chemically annotated and searchable in SureChEMBL
SureChEMBL provides search access to all patents (not just chemically annotated ones)
~120M patents

10. EMBL-EBI chemistry resources
RDF and REST API interfaces
Atlas
Ligand induced transcript response
750
PDBe
Ligand structures from structurally defined protein complexes
15K
ChEBI
Nomenclature of primary and secondary metabolites. Chemical Ontology
24K
ChEMBL
Bioactivity data from literature and depositions
1.5M
SureChEMBL
Chemical structures from patent literature
~16M
3rd Party Data
ZINC, PubChem, ThomsonPharma DOTF, IUPHAR, DrugBank, KEGG, NIH NCC, eMolecules, FDA SRS, PharmGKB, Selleck, ….
~60M
Negative novelty checking
UniChem – InChI-based chemical resolver (full + relaxed ‘lenses’)
>80M
REST API Interface -

11. SureChEMBL data access I
UniChem (“Universal Compound Resolver”)
Weekly updates
Web service lookup
Connectivity search
FTP download
Quarterly updates
All SureChEMBL compounds in SDF and CSV format
Raw data
ftp://ftp.ebi.ac.uk/pub/databases/chembl/SureChEMBL/
no further filtering or manual curation

12. SureChEMBL data access II
PubChem
SureChEMBL data source
Quarterly updates
Data feed client
Creates a local replica of SureChEMBL
Updates daily
own-replica-of.html

13. Can we have everything? Cost Quality Time
Usually you can have 2 of the 3 factors
We do not have access to commercial patent chemsitry
products such as scifinder
Quality
Time

14. Common sources of errors
Small, poor quality images
OCR errors in names (OCR done by IFI). There is an OCR correction step, but cannot fix all errors
-> ‘2,6-Difluoro-Λ/-{1 -r(4-iodo-2-methylphenyl)methvn-1 H-pyrazol-3- vDbenzamide’
Reliability better for US patents due to inclusion of mol files
Missing double bonds, rings opened, failure to extract molecule from image

15. Bioactivity data extraction?
Compounds
Target/Assay
Bioactivity
I know what many of you are thinking

16. Markush structure extraction?
-alkyl
-aryl
-heteroaryl
-heterocyclyl
-cycloalkyl ….
Accurate automated markush extraction is
We welcome ideas and collaborations

17. Future plans Full compound-patent map OpenPHACTS ENSO project
Flat file ftp download
Coming in March
Regular updates
Also available in UniChem
OpenPHACTS ENSO project
Biological entity extraction and annotation
Proteins, genes and diseases
Ontology mapping and semantic integration

18. SureChEMBL Interface

19. Homepage Filter by authority (US, EP, WO and JP) Search by keyword
Filter by document section (title, claims, abstract, description and images)
Chemical search type filter (substructure, similarity, identical)
Filter by date
Filter by MW
Search by keyword
Search by chemical structure
(sketch compound)
Search by SMILES, MOL, SMARTS, name
Search by patent number
Help

20. Keyword-based search Uses Boolean operators and Lucene query fields
Example searches…
roche OR novartis
sterili?e
kinase*
pfizer C07D “kinase inhibitor”
pn:WO A1
pa:(bayer OR genentech OR merck) AND desc:(chemotherap* AND (“phosphoinositide kinase”~0.8 OR Pi3K))
Logical operators – UPPER case
Pn patent number
Pa assignee or applicant - all lucene lower case
? represents one and only one character
* represents 0 to n characters
Quotes will retrieve documents where the phrase exists exactly as entered
TO FIND WORDS THAT ARE WITHIN A SPECIFIC PROXIMITY TO ONE ANOTHER, USE THE TILDE "~" SYMBOL AT THE END OF A PHRASE
Complex searches can be formed

21. Lucene Field
Description
Indexed Data
Sample
scpn
SureChEMBL Patent Number (SCPN)
EP B1
scpn:EP B1 pn
publication number
EP B1
pn:ep b1 pd
publication date
pd: an
application number
EP A
an:EP A ad
application date
ad:
pri
priority(ies)
DE A
pri:“DE A ”
pridate
all priority dates
pridate:
pdyear
publication year
2013
pdyear:2013 ds
designated states DE
ds:(DE OR GB OR FR) GB
ds:FR
pctpn
PCT publication number
WO A2
pctpn:WO A2
pctpd
PCT publication date
pctpd:
pctan
PCT application number
US W
pctan:US W
pctad
PCT application date
pctad:
relan
related application number
Division of application No. 12/159,232
relan:US
relad
related application date
Jun 26, 2008
relad: ic
IPCR
C CO8 C08K C08K0005
ic:C
cpc
CPC
C C07 C07D C07D0471 C07D047104
cpc:C07D
ecla
ECLA
C07D487/10
ecla:C07D487/10 uc
US class 29
uc:029
inv
inventor(s)
schmidt hans-werner
inv:("schmidt hans" AND thelakkat)
apl
applicant
Sony International (Europe) GmbH
apl:sony
asg
assignee
SIEMENS AKTIENGESELLSCHAFT
asg:siemens
pa apl or asg
assignee(s) or applicant(s)
see apl and asg above
pa:sony
cor
correspondent
Dr Roger Brooks
cor: “Dr Roger Brooks”
agt
agents
Pohlman, Sandra M
agt:”Pohlman, Sandra M”
pcit
patent citations
EP B1
pcit:EP B1
ncit
non-patent citations
TANG C W: ”Two-layer organic photovoltaic cell”
ncit:(tang AND ”Two-layer organic photovoltaic cell”)
ttl
title in English, French and German
Sonnenenergiesystem
ttl:(”solar energy” OR “énergie solaire” OR Sonnen*) ab
abstract in English, French and German
desc
description in English, French and German
clm
claims in English, French and German
text
abstract or description or claims in English, French or German
pnlang
publication language
EN FR DE PT NO RU NL SV FI TR IS and more
pnlang:(NO OR FI OR SV)

22. Fielded keyword search
Filter by document section
Lucene Field Name
pn: publication number
pa: assignee(s) or applicant(s)
Logical operators

23. SureChEMBL Patent Numbers (SCPN)
Standardised format used to search system
Format: CC-PATNO-KK, e.g. WO A2
Batch conversion available via interface homepage link
Country code
Number
Kind code

24. Keyword searches return documents

25. Patent family members

26. Export patent chemistry
Property range filters
Count filters
Go to ‘My Exports’ to download CSV or XML

27. Patent view - Front page

28. Patent view - Claims

29. Chemical entities in patent
Click on blue highlighted text to see chemical info box

30. Patent view - Tools Export chemistry for document or family
Access to source document PDF

31. Chemistry-based searching
Types of search
Structure sketch
(2 sketchers)
Filter by MW range
Different sketchers available
Substructure
Chemists are most often interested in Substructure search, that is, whether a target structure contains the query structure within it.
Note: If special molecular features are present on the query (eg. stereochemistry, charge, etc.), only those targets containing the feature are considered hits. However, if a feature is missing from the query, it is not checked and targets without that feature may appear as hits.
Similarity
A Similarity structure search looks for target structures that are similar to the query structure. The similarity concept implemented is based on hashed binary chemical fingerprints derived using a Tanimoto metric. That is to say, the presence of molecular features are recorded for both the query and the target and then compared using a standard formula.
Note: See Tanimoto Coefficient and Fingerprint Generation for a complete description of the concept and method.
Note: If you choose Similarity as your search type, you will be prompted to provide a Tanimoto coefficient between 0.5 and 0.95.
Major Match A Major Match structure search finds molecules that are equal in size to the query structure with additional fragments or heavy atoms allowed. This search type is useful to perform as it ignores the presence of salts or solvents beside the main structure in the target.
Basic A Basic structure search finds molecules that are equal (in size) to the query structure. No additional fragments or heavy atoms are allowed. By default, molecular features are evaluated the same way as described above for substructure search.
Identical In an Identical structure search, all molecular features need to be equal (e.g. a non-stereo query will only match a non-stereo target).
Filter by document section

32. Chemistry searches return structures
Tautomers are registered as different structures, unlike in ChEMBL – this will likely change in future

33. Review chemistry hits

34. Compound report page
On-the-fly integration with 81M structures and from 28 data sources
UniChem integration: On-the-fly integration with ~81M structures and from 28 data sources

35. Review patent documents for chemistry

36. Review patent documents for chemistry
families

37. SureChEMBL knowledge base

38. SureChEMBL support

39. ChEMBL blog

40. Summary SureChEMBL Interface demo Coverage and content Capabilities
Future plans
Interface demo
Surechembl free source
Complements chembl with chemistry from patent corpus
There are developments

41. Acknowledgements ChEMBL team Digital Science Open PHACTS consortium
John Overington
Jon Chambers
George Papadatos
Mark Davies
Nathan Dedman
Anna Gaulton
Digital Science
Nicko Goncharoff
James Siddle
Richard Koks
Open PHACTS consortium
Funding:
Innovative Medicines Initiative Joint Undertaking, grant agreement no (Open PHACTS)
Wellcome Trust Strategic Award for Chemogenomics, WT086151/Z/08/Z
European Molecular Biology Laboratory
European Commission FP7 Capacities Specific Programme, grant agreement no (BioMedBridges)
Software:

42. Future webinars:
25th March - UniProt: Exploring protein sequence and functional information
8th April Introduction to ENA
22nd April Ensembl Tools
6th May Reactome: Exploring biological pathways
All 4:00pm GMT
For details see: ebi-training-webinar-series-2015

43. myChEMBL Example

44. What is myChEMBL? A Virtual Machine, preloaded with…
A complete version of the ChEMBL database
Chemical structure searching
GUI & web services for accessing the database
A suite of chemoinformatics and data analysis tools
Tutorials on a range of topics
Using ChEMBL data
Chemoinformatics, machine learning, etc.
Completely free and open

45. myChEMBL: Applications
Centralised Resource
VM shareable across the local network
Access to standardised tools, services and data
Application Development
Sandboxed VM, all source code available
Learning
Lowers ‘activation barrier’ with pre-installed tools and examples
Teaching, Training & Dissemination
IPython notebooks and KNIME
2nd Prize at ACS Teach-Discover-Treat competition
Open Data combined with Open Tools on an Open OS
All-in-one platform for
learn and practice chemoinformatics
A centralised machine for chemoinformatics services
Share the VM across the local network
Local access to standardised tools, services and resources
myChEMBL uses exclusively free and open source tools and libraries, so it removes the expensive licensing costs often associated with similar applications.
myChEMBL runs locally behind a firewall, therefore the typical concerns regarding the submission of sensitive data to web-based applications do not apply.
the source code is available for all myChEMBL applications, so developers can use this as a starting point for applications they wish to develop in the future.
Due to the availability of interactive, web- and GUI-based tools, myChEMBL requires no prior programming experience or knowledge.
myChEMBL provides a versatile platform for learning chemical data mining and cheminformatics in an intuitive and straightforward way. The combination of data with relevant pre-installed tools effectively lowers the 'activation barrier' and shifts the focus to hands-on programming and learning.
Proven in our annual Wellcome Trust courses
- myChEMBL is a proven resource for training scientists on the use of essential tools in the field of chemoinformatics and computer-aided drug discovery.

46. myChEMBL LaunchPad

47. SureChEMBL and myChEMBL
More:
Download: ftp://ftp.ebi.ac.uk/pub/databases/chembl/VM/myChEMBL/current/

48. SureChEMBL and myChEMBL