Organometallics InChIng forwards to better representations and happier chemists Ian Bruno (CCDC) Colin Batchelor (RSC) Jonathan Goodman (University of Cambridge) Gerd Blanke (StructurePendium Technologies GmbH) Alex Clark (Molecular Materials Informatics)
32
Embed
Organometallics - Cambridge Structural Database€¦ · Organometallics InChIng forwards to better representations and happier chemists Ian Bruno (CCDC) Colin Batchelor (RSC) Jonathan
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
OrganometallicsInChIng forwards to better representations and happier chemists
Ian Bruno (CCDC)
Colin Batchelor (RSC)
Jonathan Goodman (University of Cambridge)
Gerd Blanke (StructurePendium Technologies GmbH)
Alex Clark (Molecular Materials Informatics)
2
Catalysing Chemistry
Unusual Highly Regioselective Direct Aldol Additions with a Moisture-Resistant and Highly Efficient Titanium CatalystMahrwald and Schetter (2005), Org. Lett., 8(2), 281https://doi.org/10.1021/ol052637z
[Rh2(COD)2(Dppm)(μ2-Cl)]BF4: Precursor for a Selective Hydrogenation Catalyst and Its Recycling by Silica EntrapmentLorenzini et al. (2006), Organometallics, 25(16), 3912https://doi.org/10.1021/om060172q
• Nearly half of the reactions in the NextMove patent reaction database include a metal
• There are probably a lot more people working in pharmaceuticals and fine chemicals who regularly use organometallics than there are specialist organometallic chemists
Source: Jonathan Goodman
3
Bioorganometallic Chemistry
RU7
PDB Chemical Component RU7 bound to PDB 1T3P
By Smokefoot - Own work, CC BY-SA 4.0https://commons.wikimedia.org/w/index.php?curid=37239342
Structure of the FeMocofactor of nitrogenase
PDB Chemical Component CFN bound to PDB 3K1A
HEM
PDB Chemical Component HEM bound to PDB 19HC
4
Functional Materials
Semi-conductors
Electrical Conductivity in Platinum-Dimer ColumnsGijarro et al. (2008) Inorg. Chem, 47(21), 9736https://doi.org/10.1021/ic801211m
Photoswitches
Understanding solid-state photoswitching in [Re(OMe2-bpy)(CO)3(η1-NO2)] crystals via in situ photocrystallographyLauren Hatcher (2018) CrystEngComm, 20, 5990-5997https://doi.org/10.1039/C8CE00774H
5
CSD:PUWTEG
CSD:GOBHIPCSD:ICUQED CSD:BOLFOY CSD:DIBFAY
CSD:WEKLIICSD:ACAROOCSD:BAHMEC
6
The Cambridge Structural Database
❑ Over 1 Million small-molecule crystal structures
❑ Over 80,000 datasets deposited annually
❑ Enriched and annotated by experts
❑ Each CSD entry includes a 2D chemical representation of the substance studied
250,000th – IBEZUK
Conducting metal-dinitrogen polymer
1,000,000th – XOPCAJ
Chalcogen bonding
catalyst
Organic43%
Not Organic57%
At least one transition metal, lanthanide, actinide or any of Al, Ga,
In, Tl, Ge, Sn, Pb, Sb, Bi, Po
CSD V5.38 Organic: *BTEST 57
7
Challenges Representing Organometallics
• Bond Perception
• Author Expression
• Stereochemistry
• Digital Representations
8
When is a Bond not a Bond?
AFEGEY: Du & Li, Acta Crystallogr.,Sect.E (2007), 63, m2282doi:10.1107/S1600536807037804
a 7.3754(18) b 8.925(2) c 13.812(3); a 92.347(2) b 97.596(2) g 96.311(3)
AFEGEY01: Seidel et al., Z.Anorg.Allg.Chem. (2011), 637, 1545, doi:10.1002/zaac.201100181
a 7.3025(4) b 8.8332(5) c 13.7017(8) a 92.919(1) b 97.319(1) g 96.456(1)
• Stereochemistry – atom and bond• Isotopic and Fixed Hydrogens (tautomerism)• Charge
13
InChI & Organometallics
GOAL: Reliable standard InChI representations of compounds in the Cambridge Structural Database (CSD) to enable intersection of this with other data resources
OUTCOME: Could confidently generate reliable InChIs for 22% of a subset of 495,751 entries from CSD 5.36
22%
58%
13%6%
2%
Order of filtering:• Not organic• Multi-component• Inchi alerts• Stereochemistry Issues
Success rate for organic portion was 51%.
Multi-component entries were excluded by choice – if all these converted successfully then success rate would be ~80% of organics.
Biggest Limitation: inability to reliably generate an InChI for metal-containing compounds
14https://iupac.org/projects/
15
Frequent Discussion Points
• Disconnection• Currently, standard InChI disconnects bonds to metal then normalises
• It is possible to generate a non-standard InChI with a reconnection layer
• Do we need to disconnect? Can we exploit existing reconnection layer?
• Stereochemistry• Do we need to be able to work out CIP priorities?
• Do we need to be able to distinguish all stereochemistries?
• External challenges• Chemists are not consistent in how they draw metal-organic structures
• Tools may not internally represent metal-organic structures the same way
• Identical structures may end up represented different ways in MOL files
16
InChI Workshop, Boston, August 2018
• Questions considered:
• What positive transformation could organometallic InChI enable?
• What are the motivational use cases?
• Organometallic InChI drivers:
• Organometallic InChI fundamental to integrity and perception of InChI: reputational risk if left unaddressed – possible barrier to adoption
• Community don't appear to be beating down the door:but perhaps patiently waiting for the working group to deliver
• Potentially significant opportunities for application:but need something to be built to explore if this is the case
17
InChI Workshop, Cambridge, February 2019
• Discussion of a Proposal from Jonathan Goodman for an Organometallic InChI Layer