Statistics, damned statistics and nanoscience - using data science to meet the challenge of nanomaterial complexity

Quasi-SMILES for predicting toxicity of Nano-mixtures to Daphnia Magna

Quasi-SMILES is an extension of the traditional SMILES. The classic SMILES is a way to represent the molecular structure. The quasi-SMILES is a way to describe all eclectic conditions that are able to affect the activity of a substance or a mixture. Nano-QSAR for prediction of toxicity of Nano-mixtures built up using the database on the corresponding experimental data. Models built up for five random splits of available data in training and validation sets are suggested. The Monte Carlo method of optimization is applied to calculate so-called optimal descriptors. The optimization was carried out with two criteria of predictive potential. These are the so-called index of ideality of correlation (IIC) and correlation intensity index (CII). Applying CII gives the better statistical quality of models for toxicity Nano-mixtures towards Daphnia Magna. The statistical quality of the best model follows the determination coefficients 0.987 (training set) and 0.980 (validation set).

Safety-by-design using forward and inverse multi-target machine learning

The economic and social future of nanotechnology depends on our ability and manufacture nanomaterials that avoid potential toxicity, by identifying them before they are made, used and released into the environment. Safety-by-design is a framework for including these issues at an early stage of the development process, but balancing multiple nanoparticle properties and selection criteria remains challenging. Based on a synthetic data set of over 19,000 possible sunscreen product specifications, we have used multi-target machine learning to predict the corresponding size, shape, concentration and polytype of titania nanoparticle additives. The study considers the optical properties responsible for the sun protection factor and product transparency, including the extinction coefficients for ultra violet and visible light, and the potential for toxicity due to the generation of reactive oxygen species from the photocatalytically active facets of both anatase and rutile nanoparticles, as a function of the size and shape. We predict a number of conventional forward structure/property and property/product relationships, but show that a direct structure/product relationship provides superior performance when predicting multiple properties or product specifications simultaneously. These models are then inverted, re-optimized and re-trained to provide focused, high performing inverse design models that do not require additional optimization, and are capable of identifying nanoparticle configurations outside of the training set. The ability to directly predict suitable nanoparticle structures that conform to prerequisite sun protection, transparently and potential toxicity thresholds represents a new approach to safety-by-design that can be applied to other products and materials where multiple design criteria must be met at the same time.

Data-Driven Materials Innovation and Applications

Owing to the rapid developments to improve the accuracy and efficiency of both experimental and computational investigative methodologies, the massive amounts of data generated have led the field of materials science into the fourth paradigm of data-driven scientific research. This transition requires the development of authoritative and up-to-date frameworks for data-driven approaches for material innovation. A critical discussion on the current advances in the data-driven discovery of materials with a focus on frameworks, machine-learning algorithms, material-specific databases, descriptors, and targeted applications in the field of inorganic materials is presented. Frameworks for rationalizing data-driven material innovation are described, and a critical review of essential subdisciplines is presented, including: i) advanced data-intensive strategies and machine-learning algorithms; ii) material databases and related tools and platforms for data generation and management; iii) commonly used molecular descriptors used in data-driven processes. Furthermore, an in-depth discussion on the broad applications of material innovation, such as energy conversion and storage, environmental decontamination, flexible electronics, optoelectronics, superconductors, metallic glasses, and magnetic materials, is provided. Finally, how these subdisciplines (with insights into the synergy of materials science, computational tools, and mathematics) support data-driven paradigms is outlined, and the opportunities and challenges in data-driven material innovation are highlighted.

Impact of atomistic or crystallographic descriptors for classification of gold nanoparticles

Machine learning models are known to be sensitive to the features used to train them, but there is currently no way to predict the impact of using different features prior to feature extraction. This is particularly important to fields such as nanotechnology that are highly multi-disciplinary, and samples can be characterised many different ways depending on the preferences of individual researchers. Does it matter if nanomaterials are described using the interatomic coordinations or more complex order parameters? In this study we compare results of supervised and unsupervised learning on a single set of gold nanoparticles that has been characterised by two different descriptors, each with a unique feature space. We find that there are some consistencies, and model selection is descriptor-agnostic, but the level of detail and the type of information that can be extracted from the results is sensitive to the way the particles are described. Unsupervised clustering revealed that an atomistic descriptor provides a finer-grained interpretation and clusters that are sub-clusters of a more sophisticated crystallographic descriptor, which is consistent with both how the features were calculated, and how they are interpreted in the domain. A supervised classifier revealed that the types of features responsible for the separation are related to the bulk structure, regardless of the descriptor, but capture different types of information. For both the atomistic and crystallographic descriptor the gradient boosting decision tree classifier gave superior results of F1-scores of 0.96 and 0.98, respectively, with excellent precision and recall, even though the clustering presented a challenging multi-classification problem.

Effect of the Albumin Corona on the Toxicity of Combined Graphene Oxide and Cadmium to Daphnia magna and Integration of the Datasets into the NanoCommons Knowledge Base

In this work, we evaluated the effect of protein corona formation on graphene oxide (GO) mixture toxicity testing (i.e., co-exposure) using the Daphnia magna model and assessing acute toxicity determined as immobilisation. Cadmium (Cd2+) and bovine serum albumin (BSA) were selected as co-pollutant and protein model system, respectively. Albumin corona formation on GO dramatically increased its colloidal stability (ca. 60%) and Cd2+ adsorption capacity (ca. 4.5 times) in reconstituted water (Daphnia medium). The acute toxicity values (48 h-EC50) observed were 0.18 mg L-1 for Cd2+-only and 0.29 and 0.61 mg L-1 following co-exposure of Cd2+ with GO and BSA@GO materials, respectively, at a fixed non-toxic concentration of 1.0 mg L-1. After coronation of GO with BSA, a reduction in cadmium toxicity of 110 % and 238% was achieved when compared to bare GO and Cd2+-only, respectively. Integration of datasets associated with graphene-based materials, heavy metals and mixture toxicity is essential to enable re-use of the data and facilitate nanoinformatics approaches for design of safer nanomaterials for water quality monitoring and remediation technologies. Hence, all data from this work were annotated and integrated into the NanoCommons Knowledge Base, connecting the experimental data to nanoinformatics platforms under the FAIR data principles and making them interoperable with similar datasets.

Exploring Two-Dimensional Materials Thermodynamic Stability via Machine Learning

The increasing interest and research on two-dimensional (2D) materials has not yet translated into a reality of diverse materials applications. To go beyond graphene and transition metal dichalcogenides for several applications, suitable candidates with desirable properties must be proposed. Here we use machine learning techniques to identify thermodynamically stable 2D materials, which is the first essential requirement for any application. According to the formation energy and energy above the convex hull, we classify materials as having low, medium, or high stability. The proposed approach enables the stability evaluation of novel 2D compounds for further detailed investigation of promising candidates, using only composition properties and structural symmetry, without the need for information about atomic positions. We demonstrate the usefulness of the model generating more than a thousand novel compounds, corroborating with DFT calculations the classification for five of these materials. To illustrate the applicability of the stable materials, we then perform a screening of electronic materials suitable for photoelectrocatalytic water splitting, identifying the potential candidate Sn₂SeTe generated by our model, and also PbTe, both not yet reported for this application.

Predicting structure/property relationships in multi-dimensional nanoparticle data using t-distributed stochastic neighbour embedding and machine learning

Combining researchers' domain expertise and advanced dimension reduction methods we demonstrate how visually comparing the distribution of nanoparticles mapped from multiple dimensions to a two dimensional plane can rapidly identify possible single-structure/property relationships and to a lesser extent multi-structure/property relationships. These relationships can be further investigated and confirmed with machine learning, using genetic programming to inform the choice of property-specific models and their hyper-parameters. In the case of our nanodiamond case study, we visually identify and confirm a strong relationship between the size and the probability of observation (stability) and a more complicated (and visually ambiguous) relationship between the ionisation potential and band gaps with a range of different structural, chemical and statistical surface features, making it more difficult to engineer in practice.

Nanoinformatics, and the big challenges for the science of small things

The combination of computational chemistry and computational materials science with machine learning and artificial intelligence provides a powerful way of relating structural features of nanomaterials with functional properties. However, combining these fundamentally different scientific approaches is not as straightforward as it seems. Machine learning methods were developed for large data sets with small numbers of consistent features. Typically nanomaterials data sets are small, with high dimensionality and high variance in the feature space, and suffer from numerous destructive biases. None of the established data science or machine learning methods in widespread use today were devised with (nano)materials data sets in mind, but there are ways to overcome these challenges and use them reliably. In this review we will discuss domain-specific constraints on data-driven nanomaterials design, and explore the differences between nanomaterials simulation and nanoinformatics that can be leveraged for greater impact.

Threshold reduction and yield improvement of semiconductor nanowire lasers via processing-related end-facet optimization

Both gain medium design and cavity geometry are known to be important for low threshold operation of semiconductor nanowire lasers. For many applications nanowire lasers need to be transferred from the growth substrate to a low-index substrate; however, the impact of the transfer process on optoelectronic performance has not been studied. Ultrasound, PDMS-assisted and mechanical rubbing are the most commonly used methods for nanowire transfer; each method may cause changes in the fracture point of the nanowire which can potentially affect both length and end-face mirror quality. Here we report on four common approaches for nanowire transfer. Our results show that brief ultrasound and PDMS-assisted transfer lead to optimized optoelectronic performance, as confirmed by ensemble median lasing threshold values of 98 and 104 μJ cm^-2 respectively, with nanowires transferred by ultrasound giving a high lasing yield of 72%. The mean threshold difference between samples is shown to be statistically significant: while a significant difference in mean length from different transfer methods is seen, it is shown by SEM that end-facet quality is also affected and plays an important role on threshold gain for this nanowire architecture.

Fluorescence Color by Data-Driven Design of Genomic Silver Clusters

DNA nucleobase sequence controls the size of DNA-stabilized silver clusters, leading to their well-known yet little understood sequence-tuned colors. The enormous space of possible DNA sequences for templating clusters has challenged the understanding of how sequence selects cluster properties and has limited the design of applications that employ these clusters. We investigate the genomic role of DNA sequence for fluorescent silver clusters using a data-driven approach. Employing rapid parallel silver cluster synthesis and fluorimetry, we determine the fluorescence spectra of silver cluster products stabilized by 1432 distinct DNA oligomers. By applying pattern recognition algorithms to this large experimental data set, we discover certain DNA base patterns, or "motifs," that correlate to silver clusters with similar fluorescence spectra. These motifs are employed in machine learning classifiers to predictively design DNA template sequences for specific fluorescence color bands. Our method improves selectivity of templates by 330% for silver clusters with peak emission wavelengths beyond 660 nm. The discovered base motifs also provide physical insights into how DNA sequence controls silver cluster size and color. This predictive design approach for color of DNA-stabilized silver clusters exhibits the potential of machine learning and data mining to increase the precision and efficiency of nanomaterials design, even for a soft-matter-inorganic hybrid system characterized by an extremely large parameter space.

Predicting the impact of structural diversity on the performance of nanodiamond drug carriers

Diamond nanoparticles (nanodiamonds) are unique among carbon nanomaterials, and are quickly establishing a niché in the biomedical application domain. Nanodiamonds are non-toxic, amenable to economically viable mass production, and can be interfaced with a variety of functional moieties. However, developmental challenges arise due to the chemical complexity and structural diversity inherent in nanodiamond samples. Nanodiamonds present a narrow, but significant, distribution of sizes, a dizzying array of possible shapes, and a complicated surface containing aliphatic and aromatic carbon. In the past these facts have been cast as hindrances, stalling development until perfectly monodispersed samples could be achieved. Current research has moved in a different direction, exploring ways that the polydispersivity of nanodiamond samples can be used as a new degree of engineering freedom, and understanding the impact our limited synthetic control really has upon structure/property relationships. In this review a series of computational and statistical studies will be summarised and reviewed, to characterise the relationship between chemical complexity, structural diversity and the reactive performance of nanodiamond drug carriers.

The impact of size and shape distributions on the electron charge transfer properties of silver nanoparticles

Many applications of silver nanoparticles are moderated by the electron charge transfer properties, such as the ionization potential, electron affinity and Fermi energy, which may be tuned by controlling the size and shape of individual particles. However, since producing samples of silver nanoparticles that are perfectly monodispersed in terms of both size and shape can be prohibitive, it is important to understand how these properties are impacted by polydispersivity, and ideally be able to predict the tolerance for variation of different geometric features. In this study, we use straightforward statistical methods, together with electronic structure simulations, to predict the electron charge transfer properties of different types of ensembles of silver nanoparticles and how restricting the structural diversity in different ways can improve or retard performance. In agreement with previous reports, we confirm that restricting the shape distribution will tune the charge transfer properties toward specific reactions, but by including the quality factors for each case we go beyond this assessment and show how targeting specific classes of morphologies and restricting the distribution of size can impact sensitivity.

Impact of distributions on the archetypes and prototypes in heterogeneous nanoparticle ensembles

The magnitude and complexity of the structural and functional data available on nanomaterials requires data analytics, statistical analysis and information technology to drive discovery. We demonstrate that multivariate statistical analysis can recognise the sets of truly significant nanostructures and their most relevant properties in heterogeneous ensembles with different probability distributions. The prototypical and archetypal nanostructures of five virtual ensembles of Si quantum dots (SiQDs) with Boltzmann, frequency, normal, Poisson and random distributions are identified using clustering and archetypal analysis, where we find that their diversity is defined by size and shape, regardless of the type of distribution. At the complex hull of the SiQD ensembles, simple configuration archetypes can efficiently describe a large number of SiQDs, whereas more complex shapes are needed to represent the average ordering of the ensembles. This approach provides a route towards the characterisation of computationally intractable virtual nanomaterial spaces, which can convert big data into smart data, and significantly reduce the workload to simulate experimentally relevant virtual samples.

Impact of speciation on the electron charge transfer properties of nanodiamond drug carriers

Unpassivated diamond nanoparticles (bucky-diamonds) exhibit a unique surface reconstruction involving graphitization of certain crystal facets, giving rise to hybrid core-shell particles containing both aromatic and aliphatic carbon. Considerable effort is directed toward eliminating the aromatic shell, but persistent graphitization of subsequent subsurface-layers makes perdurable purification a challenge. In this study we use some simple statistical methods, in combination with electronic structure simulations, to predict the impact of different fractions of aromatic and aliphatic carbon on the charge transfer properties of the ensembles of bucky-diamonds. By predicting quality factors for a variety of cases, we find that perfect purification is not necessary to preserve selectivity, and there is a clear motivation for purifying samples to improve the sensitivity of charge transfer reactions. This may prove useful in designing drug delivery systems where the release of (selected) drugs needs to be sensitive to specific conditions at the point of delivery.