Chemical Workflows Research Articles

Deductive machine learning models for product identification.

Deductive solution strategies are required in prediction scenarios that are under determined, when contradictory information is available, or more generally wherever one-to-many non-functional mappings occur. In contrast, most contemporary machine learning (ML) in the chemical sciences is inductive learning from example, with a fixed set of features. Chemical workflows are replete with situations requiring deduction, including many aspects of lab automation and spectral interpretation. Here, a general strategy is described for designing and training machine learning models capable of deduction that consists of combining individual inductive models into a larger deductive network. The training and testing of these models is demonstrated on the task of deducing reaction products from a mixture of spectral sources. The resulting models can distinguish between intended and unintended reaction outcomes and identify starting material based on a mixture of spectral sources. The models also perform well on tasks that they were not directly trained on, like performing structural inference using real rather than simulated spectral inputs, predicting minor products from named organic chemistry reactions, identifying reagents and isomers as plausible impurities, and handling missing or conflicting information. A new dataset of 1 124 043 simulated spectra that were generated to train these models is also distributed with this work. These findings demonstrate that deductive bottlenecks for chemical problems are not fundamentally insuperable for ML models.

A deep neural network for molecular wave functions in quasi-atomic minimal basis representation.

The emergence of machine learning methods in quantum chemistry provides new methods to revisit an old problem: Can the predictive accuracy of electronic structure calculations be decoupled from their numerical bottlenecks? Previous attempts to answer this question have, among other methods, given rise to semi-empirical quantum chemistry in minimal basis representation. We present an adaptation of the recently proposed SchNet for Orbitals (SchNOrb) deep convolutional neural network model [K. T. Schütt et al., Nat. Commun. 10, 5024 (2019)] for electronic wave functions in an optimized quasi-atomic minimal basis representation. For five organic molecules ranging from 5 to 13 heavy atoms, the model accurately predicts molecular orbital energies and wave functions and provides access to derived properties for chemical bonding analysis. Particularly for larger molecules, the model outperforms the original atomic-orbital-based SchNOrb method in terms of accuracy and scaling. We conclude by discussing the future potential of this approach in quantum chemical workflows.

Subnanoliter precision piezo pipette for single-cell isolation and droplet printing

Although microliter-scale liquid handling with a handheld pipette is a routine task, pipetting nanoliter-scale volumes is challenging due to several technical difficulties including surface tension, adhesion and evaporation effects. We developed a fully automated piezoelectric micropipette with a precision of < 1 nanoliter, improving the efficiency of imaging-based single-cell isolation to above 90%. This improvement is crucial when sorting rare or precious cells, especially in medical applications. The compact piezoelectric micropipette can be integrated into various (bio)chemical workflows. It eliminates plastic tubes, valves, syringes, and pressure tanks. For high-quality phase-contrast illumination of the sample, e.g., cells or tiny droplets, we constructed rings of LEDs arranged concentrically to the micropipette. The same device can be readily used for single-cell printing and nanoliter-scale droplet printing of reagents using either fluorescent or transparent illumination on a microscope. We envision that this new technology will shortly become a standard tool for single-cell manipulations in medical diagnostics, e.g., circulating tumor cell isolation.

AARON: An Automated Reaction Optimizer for New Catalysts.

We describe an open-source computational toolkit (AARON: An Automated Reaction Optimizer for New catalysts) that automates the quantum mechanical geometry optimization and characterization of the transition state and intermediate structures required to predict the activities and selectivities of asymmetric catalytic reactions. Modern computational quantum chemistry has emerged as a powerful tool for explaining the selectivity and activity of asymmetric catalysts. However, reliably predicting the stereochemical outcome of realistic reactions often requires the geometry optimization of hundreds of transition state and intermediate structures, which is a tedious process. AARON automates these optimizations through an interface with a popular electronic structure package, accelerating quantum chemical workflows to enable the computational screening of potential catalysts. AARON is built using a collection of object-oriented Perl modules (AaronTools) that provide functionality to build and modify molecular and supramolecular structures. The main functionalities of AaronTools are also available as stand-alone command-line scripts. The core features of AaronTools and AARON are explained, and representative applications of AARON to both organocatalyzed and transition-metal-catalyzed reactions are presented.

Metadata Management in the MoSGrid Science Gateway - Evaluation and the Expansion of Quantum Chemistry Support

Science gateways are employed to hide increasingly complex IT infrastructures from users via easy-to-use graphical interfaces while enabling IT-driven research not possible before. The science gateway MoSGrid (Molecular Simulation Grid) is a valuable and user-friendly workbench to submit and process molecular simulation studies on a large scale. With regard to the needs of the users, we focus on the interoperability of simulations using two prominent quantum chemical codes, Gaussian09 and NWChem. At a first glimpse, the definition of functionals and basis sets seems to be sufficient to evoke the same type of calculation in both codes using the quantum chemical workflows in MoSGrid. In more detail, this is not true and more aspects such as integration grids, convergence criteria and basis set dimensions have to be well defined in order to obtain a trustworthy comparability between quantum chemical codes. In previous work, these details have not been defined and included in the MSML (Molecular Simulation Markup Language) implementation within MoSGrid. After the investigation presented here, all these details can be integrated to extend the quantum chemical workflows in MoSGrid. Furthermore, a performance evaluation of the underlying metadata management is performed to investigate its suitability and scalability to the MSML extension.

Quantum chemical meta‐workflows in MoSGrid

SummaryQuantum chemical workflows can be built up within the science gateway Molecular Simulation Grid. Complex workflows required by the end users are dissected into smaller workflows that can be combined freely to larger meta‐workflows. General quantum chemical workflows are described here as well as the real use case of a spectroscopic analysis resulting in an end‐user desired meta‐workflow. All workflow features are implemented via Web Services Parallel Grid Runtime and Developer Environment and submitted to UNICORE. The workflows are stored in the Molecular Simulation Grid repository and ported to the SHIWA repository. Copyright © 2014 John Wiley &amp; Sons, Ltd.