Open-source tools for querying and organizing large reaction databases
3 likes2,364 views
Gregory Landrum presented on open-source tools for querying and organizing large reaction databases using the RDKit. He discussed public reaction data sources extracted from patents, handling reactions with RDKit, using fingerprints to analyze reactions, and applying machine learning and clustering to reaction fingerprints to validate their ability to distinguish reactions and group similar ones together. He also explored analyzing functional group changes between reactants and products of reactions.
![An emerging area: chemical reactions
8
Not just what we made, but how we made it
§ Text-mining applied to open patent data to extract chemical reactions :
1.12 million reactions[1]
§ Reactions classified using namerxn, when possible, into 318 standard
types : >599000 classified reactions[2]
[1] Lowe DM: “Extraction of chemical structures and reactions from the literature.” PhD
thesis. University of Cambridge: Cambridge, UK; 2012.
[2] Reaction classification from Roger Sayle and Daniel Lowe (NextMove Software)
http://nextmovesoftware.com/blog/2014/02/27/unleashing-over-a-million-reactions-into-the-
wild/
Lots of reactions,
lots of repeats](https://image.slidesharecdn.com/reactioninformatics-140501054954-phpapp02/85/Open-source-tools-for-querying-and-organizing-large-reaction-databases-8-320.jpg)
![RDKit: Some features
§ Input/Output: SMILES/SMARTS, SDF, TDT, PDB,
SLN [1], Corina mol2 [1]
§ “Cheminformatics”:
• Substructure searching
• Canonical SMILES
• Chirality support (i.e. R/S or E/Z labeling)
• Chemical transformations (e.g. remove matching
substructures)
• Chemical reactions
§ 2D depiction, including constrained depiction
§ 2D->3D conversion/conformational analysis via
distance geometry
§ UFF and MMFF94 implementation for cleaning up
structures
§ Fingerprinting: Daylight-like, atom pairs, topological
torsions, Morgan algorithm, “MACCS keys”, etc.
§ Similarity/diversity picking
§ 2D pharmacophores [1]
§ Gasteiger-Marsili charges
§ Hierarchical subgraph/fragment analysis
§ Bemis and Murcko scaffold determination
§ RECAP and BRICS implementations
§ Multi-molecule maximum common substructure
§ Feature maps
§ Shape-based similarity
§ Fraggle similarity (from GSK)
§ Molecule-molecule alignment
§ Open3DAlign implementation
§ Integration with PyMOL for 3D visualization
§ Functional group filtering
§ Salt stripping
§ Molecular descriptor library:
Topological (κ3, Balaban J, etc.), Compositional (Number
of Rings, Number of Aromatic Heterocycles, etc.),
EState, SlogP/SMR (Wildman and Crippen approach),
“MOE like” VSA descriptors, Feature-map vectors
§ Machine Learning:
• Clustering (hierarchical)
• Information theory (Shannon entropy, information
gain, etc.)
§ Tight integration with the IPython notebook and
pandas
§ Integration with the InChI library
[1] These implementations are functional but are not necessarily
the best, fastest, or most complete.](https://image.slidesharecdn.com/reactioninformatics-140501054954-phpapp02/85/Open-source-tools-for-querying-and-organizing-large-reaction-databases-11-320.jpg)
Open-source tools for querying and organizing large reaction databases
- 1. Gregory Landrum NIBR Informatics Novartis Institutes for BioMedical Research UK QSAR 2014 Open-source tools for querying and organizing large reaction databases
- 2. Outline 2 § Public data sources and reactions § Handling reactions with the RDKit § Fingerprints for reactions § Validation: • Machine learning • Clustering § Application: Identifying interesting clusters of reactions
- 3. Public data sources in cheminformatics an aside at the beginning
- 4. Protein data bank 4 the exception • Crystal structures of proteins • Deposition is mandatory for publishing protein crystal structures
- 5. Pubchem 5 Evolution Compounds Assays (non-ChEMBL) Collection of molecules from vendors and patents together with some assay data, primarily from NIH-funded screening centers.
- 6. ChEMBL 6 Evolution Compounds Activities 2009 Collection of molecules and assay data curated (primarily) from the literature
- 7. What about how we made those molecules? 7 Public reaction data? § The literature: § Plenty of data locked up in large commercial databases, very very little in the open Yan, L. et al. SAR studies of 3-arylpropionic acids as potent and selective agonists of sphingosine-1-phosphate receptor-1 (S1P1) with enhanced pharmacokinetic properties. Bioorganic & Medicinal Chemistry Letters 17, 828– 831 (2007).
- 8. An emerging area: chemical reactions 8 Not just what we made, but how we made it § Text-mining applied to open patent data to extract chemical reactions : 1.12 million reactions[1] § Reactions classified using namerxn, when possible, into 318 standard types : >599000 classified reactions[2] [1] Lowe DM: “Extraction of chemical structures and reactions from the literature.” PhD thesis. University of Cambridge: Cambridge, UK; 2012. [2] Reaction classification from Roger Sayle and Daniel Lowe (NextMove Software) http://nextmovesoftware.com/blog/2014/02/27/unleashing-over-a-million-reactions-into-the- wild/ Lots of reactions, lots of repeats
- 9. More about the classes 9 Frequency of classes, revisited: 44675 2.1.2 Carboxylic acid + amine reaction 39297 1.7.9 Williamson ether synthesis 28194 2.1.1 Amide Schotten-Baumann 26739 1.3.7 Chloro N-arylation 22400 1.6.2 Bromo N-alkylation 20465 7.1.1 Nitro to amino 20405 1.6.4 Chloro N-alkylation 17226 6.2.2 CO2H-Me deprotection 16602 6.1.1 N-Boc deprotection 16021 6.2.1 CO2H-Et deprotection 12952 1.2.1 Aldehyde reductive amination 12250 2.2.3 Sulfonamide Schotten-Baumann 10659 11.9 Separation 8538 3.1.5 Bromo Suzuki-type coupling 7261 1.7.7 Mitsunobu aryl ether synthesis 7102 6.3.7 Methoxy to hydroxy 7071 3.3.1 Sonogashira coupling 6472 3.1.1 Bromo Suzuki coupling 6383 1.8.5 Thioether synthesis 5791 9.1.6 Hydroxy to chloro 20 most common classes:
- 10. RDKit: What is it? § Open-source C++ toolkit for cheminformatics § Wrappers for Python (2.x), Java, C# § Functionality: • 2D and 3D molecular operations • Descriptor generation for machine learning • PostgreSQL database cartridge for substructure and similarity searching • Knime nodes • IPython integration • Lucene integration (experimental) • Supports Mac/Windows/Linux § Releases every 6 months § business-friendly BSD license § Code: https://github.com/rdkit § http://www.rdkit.org
- 11. RDKit: Some features § Input/Output: SMILES/SMARTS, SDF, TDT, PDB, SLN [1], Corina mol2 [1] § “Cheminformatics”: • Substructure searching • Canonical SMILES • Chirality support (i.e. R/S or E/Z labeling) • Chemical transformations (e.g. remove matching substructures) • Chemical reactions § 2D depiction, including constrained depiction § 2D->3D conversion/conformational analysis via distance geometry § UFF and MMFF94 implementation for cleaning up structures § Fingerprinting: Daylight-like, atom pairs, topological torsions, Morgan algorithm, “MACCS keys”, etc. § Similarity/diversity picking § 2D pharmacophores [1] § Gasteiger-Marsili charges § Hierarchical subgraph/fragment analysis § Bemis and Murcko scaffold determination § RECAP and BRICS implementations § Multi-molecule maximum common substructure § Feature maps § Shape-based similarity § Fraggle similarity (from GSK) § Molecule-molecule alignment § Open3DAlign implementation § Integration with PyMOL for 3D visualization § Functional group filtering § Salt stripping § Molecular descriptor library: Topological (κ3, Balaban J, etc.), Compositional (Number of Rings, Number of Aromatic Heterocycles, etc.), EState, SlogP/SMR (Wildman and Crippen approach), “MOE like” VSA descriptors, Feature-map vectors § Machine Learning: • Clustering (hierarchical) • Information theory (Shannon entropy, information gain, etc.) § Tight integration with the IPython notebook and pandas § Integration with the InChI library [1] These implementations are functional but are not necessarily the best, fastest, or most complete.
- 12. RDKit reaction handling Basics From an rxn file:
- 13. RDKit reaction handling Virtual Protecting groups The problem: Introducing the protecting group on amide Ns: The result:
- 14. Another approach for tuning specificity start with the problem again
- 15. Another approach for tuning specificity and now the solution Thanks to Holger Claussen (BioSolveIT) for the idea to use atom values for this Query definitions added as atom values
- 16. Got the reactions, what about reaction fingerprints? 16 Criteria for them to be useful § Question 1: do they contain bits that are helpful in distinguishing reactions from another? Test: can we use them with a machine-learning approach to build a reaction classifier? § Question 2: are similar reactions similar with the fingerprints Test: do related reactions cluster together?
- 17. Similarity applied to reactions 17 What are we talking about? § These two reactions are both type: “1.2.5 Ketone reductive amination” It’s obvious that these are the same, right?
- 18. Got the reactions, what about reaction fingerprints? 18 Start simple: use difference fingerprints: Similar idea here: 1) Ridder, L. & Wagener, M. SyGMa: Combining Expert Knowledge and Empirical Scoring in the Prediction of Metabolites. ChemMedChem 3, 821–832 (2008). 2) Patel, H., Bodkin, M. J., Chen, B. & Gillet, V. J. Knowledge-Based Approach to de NovoDesign Using Reaction Vectors. J. Chem. Inf. Model. 49, 1163–1184 (2009). FPReacts = FPi i∈Reactants ∑ FPProducts = FPi i∈Products ∑ FPRxn = FPProds − FPReacts
- 19. Refine the fingerprints a bit 19 Text-mined reactions often include reagents or solvents in the reactants Explore two options for handling this: 1. Decrease the weight of reactant molecules where too many of the bits are not present in the product fingerprint 2. Decrease the weight of reactant molecules where too many atoms are unmapped
- 20. Another reaction analysis scheme 20 Looking at functional group changes § Similar idea to the fingerprint analysis: count the numbers of common functional groups in the reactants and products and subtract the one from the other: rfp=None for ri in range(rxn.GetNumReactantTemplates()): m = rxn.GetReactantTemplate(ri) fp = np.array(FunctionalGroups.CreateMolFingerprint(m,fgh)) if rfp is None: rfp = fp else: rfp += fp pfp=None for ri in range(rxn.GetNumProductTemplates()): m = rxn.GetProductTemplate(ri) fp = np.array(FunctionalGroups.CreateMolFingerprint(m,fgh)) if pfp is None: pfp = fp else: pfp += fp fp = pfp-‐rfp
- 22. Functional group changes analyzed 22 Do the results make sense at all? Func%onal Group Avg in Reac%on Overall Average halogen -‐0.98 -‐0.3 alcohol -‐0.95 -‐0.12 halogen_no4luorine -‐0.89 -‐0.27 alcohol_aroma:c -‐0.67 -‐0.04 halogen_alipha:c -‐0.62 -‐0.15 halogen_no4luorine_alipha:c -‐0.62 -‐0.14 carboxylicacid -‐0.5 -‐0.23 halogen_bromine -‐0.42 -‐0.11 halogen_bromine_alipha:c -‐0.39 -‐0.06 halogen_aroma:c -‐0.36 -‐0.16 alcohol_alipha:c -‐0.28 -‐0.08 halogen_no4luorine_aroma:c -‐0.27 -‐0.13 amine -‐0.04 -‐0.3 amine_alipha:c -‐0.04 -‐0.27 carboxylicacid_alipha:c -‐0.04 -‐0.08 halogen_bromine_aroma:c -‐0.03 -‐0.05 amine_ter:ary -‐0.02 -‐0.06 amine_ter:ary_alipha:c -‐0.02 -‐0.08 carboxylicacid_aroma:c -‐0.02 -‐0.03 amine_cyclic -‐0.01 -‐0.02 halogen_bromine_bromoketone -‐0.01 0 Func%onal Group Avg in Reac%on Overall Average acidchloride 0 -‐0.07 acidchloride_alipha:c 0 -‐0.05 acidchloride_aroma:c 0 -‐0.02 aldehyde 0 -‐0.04 aldehyde_alipha:c 0 -‐0.01 aldehyde_aroma:c 0 -‐0.03 amine_aroma:c 0 -‐0.03 amine_primary 0 -‐0.15 amine_primary_alipha:c 0 -‐0.07 amine_primary_aroma:c 0 -‐0.07 amine_secondary 0 -‐0.04 amine_secondary_alipha:c 0 -‐0.07 amine_secondary_aroma:c 0 0.03 amine_ter:ary_aroma:c 0 0 azide 0 0 azide_alipha:c 0 0 azide_aroma:c 0 0 boronicacid 0 -‐0.03 boronicacid_alipha:c 0 0 boronicacid_aroma:c 0 -‐0.03 carboxylicacid_alphaamino 0 0 isocyanate 0 -‐0.01 isocyanate_alipha:c 0 0 isocyanate_aroma:c 0 0 nitro 0 -‐0.03 nitro_alipha:c 0 0 nitro_aroma:c 0 -‐0.03 sulfonylchloride 0 -‐0.02 sulfonylchloride_alipha:c 0 -‐0.01 sulfonylchloride_aroma:c 0 -‐0.01 terminalalkyne 0 -‐0.01 Compare the average deltas for the >39K instances of Williamson ether synthesis These look sensible
- 23. Are the fingerprints useful? 23 § Question 1: do they contain bits that are helpful in distinguishing reactions from another? Test: can we use them with a machine-learning approach to build a reaction classifier? § Question 2: are similar reactions similar with the fingerprints Test: do related reactions cluster together?
- 24. Machine learning and chemical reactions 24 § Validation set: • The 68 reaction types with at least 2000 instances from the patent data set - “Resolution” reaction types removed (e.g. 11.9 Separation and 11.1 Chiral separation) - Final: 66 reaction types § Process: • Training set is 200 random instances of each reaction type • Test set is 800 random instances of each reaction type • Learning: random forest (scikit-learn)
- 25. Learning reaction classes 25 Results for test data Overall: • Recall: 0.94 • Precision: 0.94 • Accuracy: 0.94 For a 66-class classifier, this looks pretty good!
- 26. Learning reaction classes 26 ~94% accuracy much of the confusion is between related types Confusion matrix for test data Bromo Suzuki coupling Bromo Suzuki-type coupling Bromo N-arylation
- 27. Are the fingerprints useful? 27 § Question 1: do they contain bits that are helpful in distinguishing reactions from another? Test: can we use them with a machine-learning approach to build a reaction classifier? § Question 2: are similar reactions similar with the fingerprints Test: do related reactions cluster together?
- 28. Clustering reactions 28 § Reaction similarity validation set: • The 66 most common reaction types from the patent data set • Look at the homogeneity of clusters with at least 10 members 1.2.5 Ketone reductive amination 1.2.5 Ketone reductive amination 1.2.5 Ketone reductive amination Integration Interpretation: <30% of clusters are <90% homogeneous Interpretation: <40% of clusters are <80% homogeneous
- 29. Similarity applied to reactions 29 Can we help classify the remaining 600K reactions? § Starting point: we have a similarity measure that clusters related reactions together § We can apply the machine-learning model to the unclassified reactions and see if the original assignment missed any instances § We can then look for big clusters of unclassified molecules and (manually) assign classes to them.
- 30. Finding related unclassified reactions 30 § Process: 1. Pick 10K random unclassified reactions 2. Cluster using the same fingerprint described above 3. Characterize clusters by average functional-group profile 4. Pick clusters where there is a clear signal § An example: Cluster 12 amine -‐0.68 amine_secondary -‐0.35 amine_secondary_aliphatic -‐0.35 amine_aliphatic -‐0.61 aldehyde -‐0.58 aldehyde_aromatic -‐0.58
- 31. Example reactions from cluster 12 31 • Clearly related reactions • Using this approach we’ve identified a number of reaction classes
- 32. Wrapping up 32 § Dataset: 1+ million reactions text mined from patents (publically available) with reaction classes assigned § Fingerprint: weighted atom-pair delta fingerprints implemented using the RDKit § Fingerprint Validation: • Multiclass random-forest classifier ~94% accurate • Similarity measure works: similar reactions cluster together § Combination of clustering + functional group analysis clustering allows identification of new reaction classes
- 33. § NIBR: • Anna Pelliccioli • Sereina Riniker • Mike Tarselli § NextMove Software: • Roger Sayle • Daniel Lowe 33 Acknowledgements
- 34. Advertising 34 3rd RDKit User Group Meeting 22-24 October 2014 Merck KGaA, Darmstadt, Germany Talks, “talktorials”, lightning talks, social activities, and a hackathon on the 24th. Registration: http://goo.gl/z6QzwD Full announcement: http://goo.gl/ZUm2wm We’re looking for speakers. Please contact greg.landrum@gmail.com