Enabling synthesis in fragment-based drug discovery (FBDD): microscale high-throughput optimisation of the medicinal chemist's toolbox reactions

Townley, Chloe; Branduardi, Davide; Chessari, Gianni; Cons, Benjamin D.; Griffiths-Jones, Charlotte; Hall, Richard J.; Johnson, Christopher N.; Ochi, Yuji; Whibley, Stuart; Grainger, Rachel

doi:10.1039/d3md00495c

Cited by 7 publications

(2 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To enable direct comparison between the various optimization approaches, we propose the expansion of the case study introduced by Shields et al to benchmark human-based decision making against three computational-based approaches, i.e., Bayesian optimization, quantum mechanics simulations, and random forest machine learning models. Specifically, we leveraged the HTE data for the Pd-catalyzed C–H arylation of N -1-methyl-1 H -imidazole-4-carbonitrile ( 1 ) with 1-bromo-2-fluorobenzene ( 2 ) (Scheme ), a common reaction type in the manufacturing of APIs. , Kinetic modeling was not considered in this exercise as only one data point was available per experiment.…”

Section: Methodologiesmentioning

confidence: 99%

“…Specifically, we leveraged the HTE data for the Pd-catalyzed C−H arylation of N-1-methyl-1H-imidazole-4carbonitrile (1) with 1-bromo-2-fluorobenzene (2) (Scheme 1), a common reaction type in the manufacturing of APIs. 43,44 Kinetic modeling was not considered in this exercise as only one data point was available per experiment. The available experimental data were complemented with an artificial "impurity" dataset calculated according to eq 2, where C max and T max are the maximum concentration and temperature of the reaction space, respectively, "yield" is the measured reaction yield provided in the HTE dataset (Shields et al 20 ), and the multiplying constant is an ad hoc factor so that the equation leads to impurities above specification in some regions of the reaction space but not on others.…”

Section: Case Study Designmentioning

confidence: 99%

See 1 more Smart Citation

Benchmarking Strategies of Sustainable Process Chemistry Development: Human-Based, Machine Learning, and Quantum Mechanics

Almeida,

Branco,

Carvalho

et al. 2024

Org. Process Res. Dev.

View full text Add to dashboard Cite

This study benchmarks diverse strategies in sustainable process chemistry development, ranging from human subject matter expertise to advanced computational models, including machine learning, Bayesian optimization, and quantum mechanics simulations. Through a "virtual laboratory" case study simulating a Pd-catalyzed C−H arylation reaction, the efficiency, sustainability, and practical application of these methodologies were compared. The study highlights the nuanced interplay between traditional expertise and computational tools, offering insights into their complementary roles in accelerating development and achieving green-by-design principles in pharmaceutical synthesis. Our findings suggest that no single approach universally outperforms others; instead, a hybrid strategy leveraging both human intuition and computational power appears to be the most promising approach when combining powerful tools in the complex field of modern organic synthesis.

show abstract

Section: Methodologiesmentioning

confidence: 99%

Section: Case Study Designmentioning

confidence: 99%

Benchmarking Strategies of Sustainable Process Chemistry Development: Human-Based, Machine Learning, and Quantum Mechanics

Almeida,

Branco,

Carvalho

et al. 2024

Org. Process Res. Dev.

View full text Add to dashboard Cite

show abstract

Palladium, [ N , N ′‐1,2‐ethanediylidenebis[2,6‐dimethylbenzenamine‐κN]][(3,4‐ η )‐2,5‐furandione] ( ^DMP DAB–Pd–MAH)

Leitch,

Thomas

2024

Encyclopedia of Reagents for Organic Synthesis

View full text Add to dashboard Cite

image [2721487‐84‐7] C 22 H 22 N 2 O 3 Pd (MW 468.85) InChI = 1S/C18H20N2.C4H2O3.Pd/c1‐13‐7‐5‐8‐14(2)17(13)19‐11‐12‐20‐18‐15(3)9‐6‐10‐16(18)4;5‐3‐1‐2‐4(6)7‐3;/h5‐12H,1‐4H3;1‐2H;/b19‐11+,20‐12+;; InChiKey = WHOQNDZHEHSRKP‐GAMUHHASSA‐N (palladium(0) compound used as a precursor to phosphine palladium(0) complexes, 1–5 and as a catalyst precursor for cross‐coupling reactions, 1,6 CH activation reactions, 7 and enantioselective allylation reactions 4,8 ) Alternative Names : (1 E ,2 E )‐ N 1, N 2‐bis(2,6‐dimethylphenyl)ethane‐1,2‐diimine 2,5‐dihydrofuran‐2,5‐dione palladium, [κ 2 ‐ N , N ′‐bis(2,6‐dimethylphenyl)ethan‐1,2‐diimine][ η 2 ‐maleic anhydride]palladium(0). Physical Data : m.p. 121 °C (dec). Solubility : fully soluble (≥20 mg mL −1 ) in acetone, dichloromethane, tetrahydrofuran (THF), 2‐methyltetrahydrofuran (MeTHF); partially soluble in toluene (5 mg mL −1 ), ethyl acetate (2.5 mg mL −1 ); poorly soluble (≤2 mg mL −1 ) in pentane, n ‐hexane, n ‐heptane, diethyl ether, tert ‐butylmethyl ether (TBME); insoluble in isopropanol (<1 mg mL −1 ); soluble but decomposes in acetonitrile, chloroform; poorly soluble and decomoposes in dimethylsulfoxide (DMSO), N,N ‐dimethylformamide (DMF), methanol, ethanol; insoluble in water. Form Supplied in : red/purple crystalline solid. Preparative Methods : commercially available. There are two reported procedures 1 for preparing DMP DAB–Pd–MAH by reacting N , N ′‐1,2‐ethanediylidenebis(2,6‐dimethylbenzenamine) ( DMP DAB) and maleic anhydride (MAH) 9 with either Pd 2 dba 3 ·CHCl 3 10 or “Pd(dba) 2 ” (actually Pd 2 dba 3 ·dba). 11 These palladium sources can be prepared from a variety of palladium(II) starting materials according to the method of Zalesskiy and Ananikov, 12 or purchased from commercial sources. Recystallization of crude “Pd(dba) 2 ” is not required, since purification of DMP DAB–Pd–MAH removes residual dba and colloidal palladium(0) (see below). N , N ′‐1,2‐ethanediylidenebis(2,6‐dimethylbenzenamine) ( DMP DAB) is prepared from glyoxal (40% wt/wt solution in water) and 2,6‐dimethylaniline in methanol with 13 or without 14 formic acid as a catalyst. Purification : colloidal palladium(0) can be removed from DMP DAB–Pd–MAH by dissolving the solid in a minimum amount of acetone, dichloromethane, or THF, followed by filtration through a bed of Celite ™ and subsequent solvent removal from the filtrate. Uncoordinated N , N ′‐1,2‐ethanediylidenebis(2,6‐dimethylbenzenamine) ( DMP DAB), maleic anhydride, and/or residual dibenzylideneacetone (dba) from the synthesis can be removed by slurrying solid DMP DAB–Pd–MAH in TBME (∼30 mL g −1 of solid), followed by filtration and washing with TBME. Analysis of Reagent Purity : 1 H NMR spectroscopic analysis is suitable for routine purity determination. In d 6 ‐acetone (solvent residual peak at 2.05 ppm), DMP DAB–Pd–MAH has diagnostic resonances at 8.55 ppm (s, 2 H ) for the N=CH–CH=N protons, 3.57 ppm (s, 2 H ) for the coordinated maleic anhydride, and 2.30 ppm (s, 12 H ) for the four Ar–CH 3 groups. Elemental analysis is also suitable for establishing bulk purity. Handling, Storage, and Precautions : solid DMP DAB–Pd–MAH is shelf‐stable when stored at room temperature in a closed container under ambient atmosphere.

show abstract

Parallel and High Throughput Reaction Monitoring with Computer Vision

Barrington,

McCabe,

Donnachie

et al. 2024

Angewandte Chemie

View full text Add to dashboard Cite

We report the development and applications of a computer vision based reaction monitoring method for parallel and high throughput experimentation (HTE). Whereas previous efforts reported methods to extract bulk kinetics of one reaction from one video, this new approach enables one video to capture bulk kinetics of multiple reactions running in parallel. Case studies, in and beyond well‐plate high throughput settings, are described. Analysis of parallel dye‐quenching hydroxylations, DMAP‐catalysed esterification, solid‐liquid sedimentation dynamics, metal catalyst degradation, and biologically‐relevant sugar‐mediated nitro reduction reactions have each provided insight into the scope and limitations of camera‐enabled high throughput kinetics as a means of widening known analytical bottlenecks in HTE for reaction discovery, mechanistic understanding, and optimisation. It is envisaged that the nature of the multi‐reaction time‐resolved datasets made available by this analytical approach will later serve a broad range of downstream efforts in machine learning approaches to exploring chemical space.

show abstract

Enabling synthesis in fragment-based drug discovery (FBDD): microscale high-throughput optimisation of the medicinal chemist's toolbox reactions

Abstract: Democratised high-throughput experimentation for FBDD.

Cited by 7 publications

References 56 publications

Benchmarking Strategies of Sustainable Process Chemistry Development: Human-Based, Machine Learning, and Quantum Mechanics

Benchmarking Strategies of Sustainable Process Chemistry Development: Human-Based, Machine Learning, and Quantum Mechanics

Palladium, [ N , N ′‐1,2‐ethanediylidenebis[2,6‐dimethylbenzenamine‐κN]][(3,4‐ η )‐2,5‐furandione] ( ^DMP DAB–Pd–MAH)

Parallel and High Throughput Reaction Monitoring with Computer Vision

Contact Info

Product

Resources

About

Enabling synthesis in fragment-based drug discovery (FBDD): microscale high-throughput optimisation of the medicinal chemist's toolbox reactions

Abstract: Democratised high-throughput experimentation for FBDD.

Cited by 7 publications

References 56 publications

Benchmarking Strategies of Sustainable Process Chemistry Development: Human-Based, Machine Learning, and Quantum Mechanics

Benchmarking Strategies of Sustainable Process Chemistry Development: Human-Based, Machine Learning, and Quantum Mechanics

Palladium, [ N , N ′‐1,2‐ethanediylidenebis[2,6‐dimethylbenzenamine‐κN]][(3,4‐ η )‐2,5‐furandione] ( DMP DAB–Pd–MAH)

Parallel and High Throughput Reaction Monitoring with Computer Vision

Contact Info

Product

Resources

About

Palladium, [ N , N ′‐1,2‐ethanediylidenebis[2,6‐dimethylbenzenamine‐κN]][(3,4‐ η )‐2,5‐furandione] ( ^DMP DAB–Pd–MAH)