Current and future machine learning approaches for modeling atmospheric cluster formation

Kubečka, Jakub; Knattrup, Yosef; Engsvang, Morten; Jensen, Andreas Buchgraitz; Ayoubi, Daniel; Wu, Haide; Christiansen, Ove; Elm, Jonas

doi:10.1038/s43588-023-00435-0

Cited by 6 publications

(5 citation statements)

References 113 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the previous section, we demonstrated that training on the binding energies of equilibrium configurations yields large errors when predicting binding energies for out-of-equilibrium structures. In Figure 6 , we show how an ML model can be tested on its transferability and extrapolation, 54 and we further examine these options in the following sections.…”

Section: Resultsmentioning

confidence: 99%

Clusterome: A Comprehensive Data Set of Atmospheric Molecular Clusters for Machine Learning Applications

et al. 2023

Self Cite

View full text Add to dashboard Cite

Formation and growth of atmospheric molecular clusters into aerosol particles impact the global climate and contribute to the high uncertainty in modern climate models. Cluster formation is usually studied using quantum chemical methods, which quickly becomes computationally expensive when system sizes grow. In this work, we present a large database of ∼250k atmospheric relevant cluster structures, which can be applied for developing machine learning (ML) models. The database is used to train the ML model kernel ridge regression (KRR) with the FCHL19 representation. We test the ability of the model to extrapolate from smaller clusters to larger clusters, between different molecules, between equilibrium structures and out-of-equilibrium structures, and the transferability onto systems with new interactions. We show that KRR models can extrapolate to larger sizes and transfer acid and base interactions with mean absolute errors below 1 kcal/mol. We suggest introducing an iterative ML step in configurational sampling processes, which can reduce the computational expense. Such an approach would allow us to study significantly more cluster systems at higher accuracy than previously possible and thereby allow us to cover a much larger part of relevant atmospheric compounds.

show abstract

Section: Resultsmentioning

confidence: 99%

Clusterome: A Comprehensive Data Set of Atmospheric Molecular Clusters for Machine Learning Applications

et al. 2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…Since this QC method is computationally quite fast, using ML does not speed up the process. More useful is, for instance, Δ-ML ωB97X‑D∥GFN1‑xTB as used in our previous work on the CS of SA-multibase clusters or even Δ-ML DLPNO∥ωB97X‑D as suggested in our recent perspective . However, as a proof of concept, we will use r 2 SCAN-3c in the next section to show the potential ML speedup by the categorization trick.…”

Section: Application and Discussionmentioning

confidence: 99%

“…The recent explosion of ML utilization in quantum chemistry has shown that ML potentials can mimic potential energy surfaces within the chemical accuracy of QC methods. There are several ML techniques for regression tasks, such as artificial neural networks (NN), Gaussian process regression (GPR), and kernel ridge regression (KRR), each with their own strengths and weaknesses . The first task in creating an ML model is choosing the molecular representation for the studied system.…”

Section: Methodsmentioning

confidence: 99%

“…This database contains clusters composed of molecules responsible for atmospheric new particle formation (NPF): acids (e.g., sulfuric and nitric), bases (e.g., ammonia and dimethylamine), water, 78 Gaussian process regression (GPR), 79 and kernel ridge regression (KRR), 80 each with their own strengths and weaknesses. 81 The first task in creating an ML model is choosing the molecular representation for the studied system. Such a representation should be invariant to transformations that do not change the particular property (translation, rotation, mirroring, or nuclear permutation).…”

Section: Usedmentioning

confidence: 99%

See 1 more Smart Citation

Computational Tools for Handling Molecular Clusters: Configurational Sampling, Storage, Analysis, and Machine Learning

Kubečka,

Besel,

Neefjes

et al. 2023

ACS Omega

Self Cite

View full text Add to dashboard Cite

Computational modeling of atmospheric molecular clusters requires a comprehensive understanding of their complex configurational spaces, interaction patterns, stabilities against fragmentation, and even dynamic behaviors. To address these needs, we introduce the Jammy Key framework, a collection of automated scripts that facilitate and streamline molecular cluster modeling workflows. Jammy Key handles file manipulations between varieties of integrated third-party programs. The framework is divided into three main functionalities: (1) Jammy Key for configurational sampling (JKCS) to perform systematic configurational sampling of molecular clusters, (2) Jammy Key for quantum chemistry (JKQC) to analyze commonly used quantum chemistry output files and facilitate database construction, handling, and analysis, and (3) Jammy Key for machine learning (JKML) to manage machine learning methods in optimizing molecular cluster modeling. This automation and machine learning utilization significantly reduces manual labor, greatly speeds up the search for molecular cluster configurations, and thus increases the number of systems that can be studied. Following the example of the Atmospheric Cluster Database (ACDB) of Elm (ACS Omega, 4, 10965–10984, 2019), the molecular clusters modeled in our group using the Jammy Key framework have been stored in an improved online GitHub repository named ACDB 2.0. In this work, we present the Jammy Key package alongside its assorted applications, which underline its versatility. Using several illustrative examples, we discuss how to choose appropriate combinations of methodologies for treating particular cluster types, including reactive, multicomponent, charged, or radical clusters, as well as clusters containing flexible or multiconformer monomers or heavy atoms. Finally, we present a detailed example of using the tools for atmospheric acid–base clusters.

show abstract

“…39,40 However, to mimic accurate DFT energies, kernel-based ML methods become computationally demanding 40–42 and neural-networks will require an extensive set of training data and hyperparameter optimization. 43 Moreover, ML methods often fail when predicting on structures different from the training set.…”

Section: Introductionmentioning

confidence: 99%

Reparameterization of GFN1-xTB for atmospheric molecular clusters: applications to multi-acid–multi-base systems

Knattrup,

Kubečka,

et al. 2024

RSC Adv.

Self Cite

View full text Add to dashboard Cite

show abstract

Current and future machine learning approaches for modeling atmospheric cluster formation

Cited by 6 publications

References 113 publications

Clusterome: A Comprehensive Data Set of Atmospheric Molecular Clusters for Machine Learning Applications

Clusterome: A Comprehensive Data Set of Atmospheric Molecular Clusters for Machine Learning Applications

Computational Tools for Handling Molecular Clusters: Configurational Sampling, Storage, Analysis, and Machine Learning

Reparameterization of GFN1-xTB for atmospheric molecular clusters: applications to multi-acid–multi-base systems

Contact Info

Product

Resources

About