A hybrid modeling approach for assessing mechanistic models of small molecule partitioning in vivo using a machine learning-integrated modeling platform

Antontsev, Victor; Jagarapu, Aditya; Bundey, Yogesh; Hou, Hypatia; Khotimchenko, Maksim; Walsh, Jason; Varshney, Jyotika

doi:10.1038/s41598-021-90637-1

Cited by 15 publications

(7 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results show that identifying the most suitable ML method and hyperparameter in the database can effectively address this issue. Moreover, advances in multiheaded attention-based transformer algorithms have been successfully implemented in numerous systems. , In addition, the state-of-art physics-informed neural networks or hybrid modeling algorithms have been shown to improve the interpretability and training efficiency for the ML models, which pioneer of ML-assisted catalysis and materials science studies. , …”

Section: Resultsmentioning

confidence: 99%

Machine-Learning-Accelerated Screening of Double-Atom/Cluster Electrocatalysts for the Oxygen Reduction Reaction

Luo,

Du,

et al. 2023

J. Phys. Chem. C

View full text Add to dashboard Cite

Carbon-based double-atom/nanocluster electrocatalysts usually demonstrate high reactivity toward the oxygen reduction reaction (ORR). However, experimental screening of optimized double-atom- and nanocluster-based ORR catalysts is often expensive and time-consuming. In this work, density functional theory (DFT) calculation is combined with the machine learning (ML) method to accelerate the screening and prediction of high-performance double-atom and nanocluster-based ORR catalysts. A database consisting of 330 ORR intermediate adsorption energies on 110 catalyst models is constructed by DFT calculations, which allows a quick ML screening of 1200 candidate ORR catalysts. The reliability of the ML model is evaluated by the R-square score (R2) and mean absolute error methods. A set of 25 potential active double-atom and nanocluster-based ORR catalysts are selected. On the basis of this ML screening, the binding energy, Bader charge transfer, and ORR reaction kinetics of the ML-predicted catalysts are considered further. The carbon-based Fe–Ce double-atom catalyst is predicted to be the best-performing ORR catalyst in the sample space. The adsorption energy-based DFT–ML framework provides an attractive approach to accelerate the screening of efficient double-atom- or cluster-based ORR electrocatalysts.

show abstract

Section: Resultsmentioning

confidence: 99%

Machine-Learning-Accelerated Screening of Double-Atom/Cluster Electrocatalysts for the Oxygen Reduction Reaction

Luo,

Du,

et al. 2023

J. Phys. Chem. C

View full text Add to dashboard Cite

show abstract

“…These algorithms were optimized to the data available in the training set. Algorithm performance in selecting essential features were proportional to the training data size [ 58 ]. BIOiSIM™’s AI infrastructure included automated statistical learning algorithms prioritizing a large collection of features and selecting those features that improved model performance.…”

Section: Methodsmentioning

confidence: 99%

Integrating AI/ML Models for Patient Stratification Leveraging Omics Dataset and Clinical Biomarkers from COVID-19 Patients: A Promising Approach to Personalized Medicine

Bello

Bundey

Bhave

et al. 2023

IJMS

Self Cite

View full text Add to dashboard Cite

The COVID-19 pandemic has presented an unprecedented challenge to the healthcare system. Identifying the genomics and clinical biomarkers for effective patient stratification and management is critical to controlling the spread of the disease. Omics datasets provide a wealth of information that can aid in understanding the underlying molecular mechanisms of COVID-19 and identifying potential biomarkers for patient stratification. Artificial intelligence (AI) and machine learning (ML) algorithms have been increasingly used to analyze large-scale omics and clinical datasets for patient stratification. In this manuscript, we demonstrate the recent advances and predictive accuracies in AI- and ML-based patient stratification modeling linking omics and clinical biomarker datasets, focusing on COVID-19 patients. Our ML model not only demonstrates that clinical features are enough of an indicator of COVID-19 severity and survival, but also infers what clinical features are more impactful, which makes our approach a useful guide for clinicians for prioritization best-fit therapeutics for a given cohort of patients. Moreover, with weighted gene network analysis, we are able to provide insights into gene networks that have a significant association with COVID-19 severity and clinical features. Finally, we have demonstrated the importance of clinical biomarkers in identifying high-risk patients and predicting disease progression.

show abstract

“…Mechanistic modelling is growing in popularity for modelling the behaviour of clinical trial outcomes [9]. In some cases, mechanistic models can be used to complement machine learning approaches [10,11] where there is both background knowledge and sufficient data available.…”

Section: Machine Learning Techniques To Generate Synthetic Datamentioning

confidence: 99%

The potential synergies between synthetic data and in silico trials in relation to generating representative virtual population cohorts

Myles¹,

Ordish²,

Tucker

2023

Prog. Biomed. Eng.

View full text Add to dashboard Cite

In silico trial methods promise to improve the path to market for both medicines and medical devices, targeting the development of products, reducing reliance on animal trials, and providing adjunct evidence to bolster regulatory submissions. In silico trials are only as good as the simulated data which underpins them, consequently, often the most difficult challenge when creating robust in silico models is the generation of simulated measurements or even virtual patients that are representative of real measurements and patients. This article digests the current state of the art for generating synthetic patient data outside the context of in silico trials and outlines potential synergies to unlock the potential of in silico trials using virtual populations, by exploiting synthetic patient data to model effects on a more diverse and representative population. Synthetic data could be defined as artificial data that mimic the properties and relationships in real data. Recent advances in synthetic data generation methodologies have allowed for the generation of high-fidelity synthetic data that are both statistically and clinically, indistinguishable from real patient data. Other experimental work has demonstrated that synthetic data generation methods can be used for selective sample boosting of underrepresented groups. This article will provide a brief outline of synthetic data generation approaches and discuss how evaluation frameworks developed to assess synthetic data fidelity and utility could be adapted to evaluate the similarity of virtual patients used for in silico trials, to real patients. The article will then discuss outstanding challenges and areas for further research that would advance both synthetic data generation methods and in silico trial methods. Finally, the article will also provide a perspective on what evidence will be required to facilitate wider acceptance of in silico trials for regulatory evaluation of medicines and medical devices, including implications for post marketing safety surveillance.

show abstract

A hybrid modeling approach for assessing mechanistic models of small molecule partitioning in vivo using a machine learning-integrated modeling platform

Cited by 15 publications

References 50 publications

Machine-Learning-Accelerated Screening of Double-Atom/Cluster Electrocatalysts for the Oxygen Reduction Reaction

Machine-Learning-Accelerated Screening of Double-Atom/Cluster Electrocatalysts for the Oxygen Reduction Reaction

Integrating AI/ML Models for Patient Stratification Leveraging Omics Dataset and Clinical Biomarkers from COVID-19 Patients: A Promising Approach to Personalized Medicine

The potential synergies between synthetic data and in silico trials in relation to generating representative virtual population cohorts

Contact Info

Product

Resources

About