<i>In silico</i>profiling nanoparticles: predictive nanomodeling using universal nanodescriptors and various machine learning approaches

Yan, Xiliang; Sedykh, Alexander; Wang, Wenyi; Zhao, Xiaoli; Yan, Bing; Zhu, Hao

doi:10.1039/c9nr00844f

Cited by 68 publications

(74 citation statements)

References 80 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning is fast becoming an important tool in the discovery and design of new materials and nanostructures. [1,2,3,4,5] It has recently been applied to a variety of applications, [6,7,8,10,11,12] including biomedical and toxicity studies. [13,14,15] Nanoinformatics is, in many ways, more challenging than it's predecessors cheminformatics and materials informatics.…”

Section: Introductionmentioning

confidence: 99%

Selecting Machine Learning Models for Metallic Nanoparticles

Barnard¹,

Opletal²

2020

Preprint

View full text Add to dashboard Cite

The outcome of machine learning is influenced by the features used to describe the data, and various metrics are used to measure model performance. In this study we use five different feature sets to describe the same 4000 gold nanoparticles, and 14 different machine learning methods to compare a total of 70 high scoring models. We then use classification and regression to show which meta-features of data sets or machine learning algorithms are important when making a selection. We find that number of features, and those that are strongly correlated, determine the class of model that should be used, but overall quality is almost entirely determined by the cross-validation score, regardless of the sophistication of the algorithm.

show abstract

Section: Introductionmentioning

confidence: 99%

Selecting Machine Learning Models for Metallic Nanoparticles

Barnard¹,

Opletal²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The Delaunay tessellation approach decomposes the nanostructure surface into tetrahedra, in which vertices are atoms. The four atoms within a tetrahedron are uniquely selected such that their circumscribing sphere does not contain any of the other atoms [250,251]. Testing 191 unique AuNP with diverse biological activities and physichochemical properties in developing QNAR models validated the suitability of the obtained novel geometrical descriptors for quantitative modeling.…”

Section: In Silico Design and Study Of Nanoparticlesmentioning

confidence: 92%

“…As the integration of computational modeling in the design of new nanomaterials progresses, there is a clear need for new universal nanodescriptors that can be used to characterize their diversity without the need for intense computational power. The applicability of the Delaunay tessellation approach, which was used in decomposing protein structures and protein-ligand bindings [248,249], to represent the nanostructures (i.e., to simulate the nanomaterial's surface chemistry) and the use of Pauling electronegativity as empirical information to define descriptor characters have been explored [250]. The Delaunay tessellation approach decomposes the nanostructure surface into tetrahedra, in which vertices are atoms.…”

Section: In Silico Design and Study Of Nanoparticlesmentioning

confidence: 99%

Understanding Nanoparticle Toxicity to Direct a Safe-by-Design Approach in Cancer Nanomedicine

et al. 2020

View full text Add to dashboard Cite

Nanomedicine is a rapidly growing field that uses nanomaterials for the diagnosis, treatment and prevention of various diseases, including cancer. Various biocompatible nanoplatforms with diversified capabilities for tumor targeting, imaging, and therapy have materialized to yield individualized therapy. However, due to their unique properties brought about by their small size, safety concerns have emerged as their physicochemical properties can lead to altered pharmacokinetics, with the potential to cross biological barriers. In addition, the intrinsic toxicity of some of the inorganic materials (i.e., heavy metals) and their ability to accumulate and persist in the human body has been a challenge to their translation. Successful clinical translation of these nanoparticles is heavily dependent on their stability, circulation time, access and bioavailability to disease sites, and their safety profile. This review covers preclinical and clinical inorganic-nanoparticle based nanomaterial utilized for cancer imaging and therapeutics. A special emphasis is put on the rational design to develop non-toxic/safe inorganic nanoparticle constructs to increase their viability as translatable nanomedicine for cancer therapies.

show abstract

“…[40] Novel "universal" descriptors for nanoparticles were also reported by Yan et al and used to generate ML models of gold nanoparticle properties using random forest and k-nearest neighbor (kNN) algorithms. [41] The descriptors were generated by Delaunay tessellation of the surface of the nanoparticles (a particular way of joining a set of points to make a triangular mesh) and summing Pauling electronegativity of atoms in each tessellation cell to represent the nanostructures (to simulate the nanomaterial's surface properties). The efficacy of these nanodescriptors was verified by modeling six gold nanoparticle datasets.…”

Section: Inadequacy Of Nanospecific Descriptors To Represent Nanomatementioning

confidence: 99%

Role of Artificial Intelligence and Machine Learning in Nanosafety

Winkler

2020

Small

View full text Add to dashboard Cite

Robotics and automation provide potentially paradigm shifting improvements in the way materials are synthesized and characterized, generating large, complex data sets that are ideal for modeling and analysis by modern machine learning (ML) methods. Nanomaterials have not yet fully captured the benefits of automation, so lag behind in the application of ML methods of data analysis. Here, some key developments in, and roadblocks to the application of ML methods are reviewed to model and predict potentially adverse biological and environmental effects of nanomaterials. This work focuses on the diverse ways a range of ML algorithms are applied to understand and predict nanomaterials properties, provides examples of the application of traditional ML and deep learning methods to nanosafety, and provides context and future perspectives on developments that are likely to occur, or need to occur in the near future that allow artificial intelligence to make a deeper contribution to nanosafety.

show abstract

In silicoprofiling nanoparticles: predictive nanomodeling using universal nanodescriptors and various machine learning approaches

Abstract: We designed novel nanodescriptors that can characterize the nanostructure diversity and also be quickly calculated in batches, to profile nanoparticles.

Cited by 68 publications

References 80 publications

Selecting Machine Learning Models for Metallic Nanoparticles

Selecting Machine Learning Models for Metallic Nanoparticles

Understanding Nanoparticle Toxicity to Direct a Safe-by-Design Approach in Cancer Nanomedicine

Role of Artificial Intelligence and Machine Learning in Nanosafety

Contact Info

Product

Resources

About