Multiscale modelling of biomolecular corona formation on metallic surfaces

• Parinaz Mosaddeghi Amini,
• Ian Rouse,
• Julia Subbotina and
• Vladimir Lobaskin

Beilstein J. Nanotechnol. 2024, 15, 215–229, doi:10.3762/bjnano.15.21

• proteins using the same fragment-based approach. To avoid the need to run a time-consuming parameterization protocol based on metadynamics simulations, we produce PMFs for the glucose bead using a machine-learning technique (PMFPredictor) trained on previous metadynamics results [38]. For the lactose

• decomposition, generating PMFs via traditional or machine-learning approaches, and constructing a coarse-grained representation for input to UA. To simplify this procedure for more complex molecules, we have developed a Python script (MolToFragments.py) employing RDKit [46] to automate splitting larger

Prediction of cytotoxicity of heavy metals adsorbed on nano-TiO₂ with periodic table descriptors using machine learning approaches

• Joyita Roy,
• Souvik Pore and
• Kunal Roy

Beilstein J. Nanotechnol. 2023, 14, 939–950, doi:10.3762/bjnano.14.77

• nanomaterials based on structure similarities with known substances. Materials with similar structures are likely to produce similar toxicity through comparable mechanisms. The development of machine learning (ML) approaches, such as artificial neural networks (ANNs), decision trees, logistic regression (LR

• model depending on the previous one. In the bagging algorithm, replica data sets are generated that minimize prediction variance in machine learning. An iterative algorithm performs a series of repeated steps to gradually improve the model’s performance or to optimize a specific parameter. The

• adjusting its parameters to minimize a specific cost or error function. These algorithms play a crucial role in training machine learning models and are fundamental to many optimization and learning techniques. Fine-tuning the model parameters through iterations helps to improve the model’s performance and

A wearable nanoscale heart sound sensor based on P(VDF-TrFE)/ZnO/GR and its application in cardiac disease detection

• Yi Luo,
• Jian Liu,
• Jiachang Zhang,
• Yu Xiao,
• Ying Wu and
• Zhidong Zhao

Beilstein J. Nanotechnol. 2023, 14, 819–833, doi:10.3762/bjnano.14.67

• energy harvesting [15]. Applying machine learning classification algorithms in the domain of human physiological signal detection is presently a prominent area of research. A notable study by R. Guo et al. [16] successfully integrated deep learning techniques with frictional hydrogel sensors to achieve

• corresponding results of the piezoelectric coefficient are presented in Figure 14b. KNN heart sound classification recognition algorithm The KNN classification algorithm is widely used in machine learning, and its fundamental classification idea is that a sample in a feature space belongs to the same category

• represents the seven optimal bases that have been selected. In this experiment, the “Classification Learner toolbox” in MATLAB was used to train a heart sound classification model. This toolbox allowed us to explore supervised machine learning by selecting various classifiers, exploring data, selecting

Transferability of interatomic potentials for silicene

• Marcin Maździarz

Beilstein J. Nanotechnol. 2023, 14, 574–585, doi:10.3762/bjnano.14.48

• –Weber, EDIP, ReaxFF, COMB, and machine-learning-based interatomic potentials. A quantitative systematic comparison and a discussion of the results obtained are reported. Keywords: 2D materials; DFT; force fields; interatomic potentials; mechanical properties; silicene; Introduction We are living in

• silicon and five polymorphs of silicon dioxide SNAP [43]: the machine-learning-based (ML-IAP) linear variant of spectral neighbor analysis potential (SNAP) fitted to total energies and interatomic forces in ground-state Si, strained structures, and slab structures obtained from DFT calculations qSNAP [43

• ]: the machine-learning-based (ML-IAP) quadratic variant of spectral neighbor analysis potential (qSNAP) fitted to total energies and interatomic forces in ground-state Si, strained structures, and slab structures obtained from DFT calculations SO(3) [44]: the machine-learning-based (ML-IAP) variant of

Frequency-dependent nanomechanical profiling for medical diagnosis

• Santiago D. Solares and
• Alexander X. Cartagena-Rivera

Beilstein J. Nanotechnol. 2022, 13, 1483–1489, doi:10.3762/bjnano.13.122

• also enable performance analyses that could be extended to more complex and relevant scenarios, aided by advanced analysis tools such as machine learning (see below in Figure 1). Decoding frequency-dependent nanomechanical measurements for disease study and follow-up The broad adoption of AFM

• latter method raises serious concerns about the patient’s health. After acquisition of tissue physical data, a computer data analysis is performed to determine the frequency-dependent mechanical properties (e.g., ES and EL). Then, an unsupervised machine learning cluster analysis is performed to identify

A superconducting adiabatic neuron in a quantum regime

• Marina V. Bastrakova,
• Dmitrii S. Pashin,
• Dmitriy A. Rybin,
• Andrey E. Schegolev,
• Nikolay V. Klenov,
• Igor I. Soloviev,
• Anastasiya A. Gorchavkina and
• Arkady M. Satanin

Beilstein J. Nanotechnol. 2022, 13, 653–665, doi:10.3762/bjnano.13.57

• obtained results indicate the conditions under which the neuron possesses the required sigmoid activation function. Keywords: Josephson junction; quantum neuron; quantum-classical neural networks; superconducting quantum interferometer; Introduction The implementation of machine learning algorithms is

• such a solution since both superconducting quantum machine learning circuits [16][17][18][19][20][21][22] and superconducting ANNs [23][24][25][26][27][28][29][30][31][32][33][34][35][36][37] are rapidly developed nowadays. Robust implementation of the considered quantum-classical system would benefit

Tunable superconducting neurons for networks based on radial basis functions

• Andrey E. Schegolev,
• Nikolay V. Klenov,
• Sergey V. Bakurskiy,
• Igor I. Soloviev,
• Mikhail Yu. Kupriyanov,
• Maxim V. Tereshonok and
• Anatoli S. Sidorenko

Beilstein J. Nanotechnol. 2022, 13, 444–454, doi:10.3762/bjnano.13.37

• broadband group signal is extremely important. Also, probabilistic analysis is used in the consideration of stochastic processes [1][2][3][4], as a popular machine learning method for spatial interpolation of non-stationary and non-Gaussian data [5], as a central part of a compensation block to enhance the

A chemiresistive sensor array based on polyaniline nanocomposites and machine learning classification

• Jiri Kroutil,
• Alexandr Laposa,
• Ali Ahmad,
• Jan Voves,
• Vojtech Povolny,
• Ladislav Klimsa,
• Marina Davydova and
• Miroslav Husak

Beilstein J. Nanotechnol. 2022, 13, 411–423, doi:10.3762/bjnano.13.34

• composed by seven different conductive sensors with composite sensing layers are measured and analyzed using machine learning. Statistical tools, such as principal component analysis and linear discriminant analysis, are used as dimensionality reduction methods. Five different classification methods

• dioxide, nitrogen dioxide, carbon monoxide, acetone, and toluene). Moreover, the obtained data were used for machine learning classification. Many pattern recognition models based on intuitive, linear and nonlinear supervised techniques have been explored in E-nose data [11][12]. A considerable number of

• increasing interpretability. We apply PCA and LDA as input data for five machine learning algorithms with a 10-fold cross-validation method. The preprocessing stage was implemented by applying PCA and LDA on the extracted dataset [14][15]. Five different kinds of flexible pattern recognition algorithms have

Identifying diverse metal oxide nanomaterials with lethal effects on embryonic zebrafish using machine learning

• Richard Liam Marchese Robinson,
• Haralambos Sarimveis,
• Philip Doganis,
• Xiaodong Jia,
• Marianna Kotzabasaki,
• Christiana Gousiadou,
• Stacey Lynn Harper and
• Terry Wilkins

Beilstein J. Nanotechnol. 2021, 12, 1297–1325, doi:10.3762/bjnano.12.97

• , embryonic zebrafish (Danio Rerio) are recognised as a key human safety relevant in vivo test system. Hence, machine learning models were developed for identifying metal oxide nanomaterials causing lethality to embryonic zebrafish up to 24 hours post-fertilisation, or excess lethality in the period of 24–120

• these modelling results are on truly external data, which were not used to select the single descriptor model. This will require further laboratory work to generate comparable data to those studied herein. Keywords: data augmentation; embryonic zebrafish; machine learning; nanosafety; nano-QSAR

• that survived at 24 hpf in both the control and dosed groups.) Hence, the modelled lethality data reported at 120 hpf, derived from the raw counts data, may be considered “excess lethality”. Two different machine learning algorithms were applied to try and learn relationships between either the 24 hpf

A review on slip boundary conditions at the nanoscale: recent development and applications

• Ruifei Wang,
• Jin Chai,
• Bobo Luo,
• Xiong Liu,
• Jianting Zhang,
• Min Wu,
• Mingdan Wei and
• Zhuanyue Ma

Beilstein J. Nanotechnol. 2021, 12, 1237–1251, doi:10.3762/bjnano.12.91

• conversion from 3 to 70% [23]. Generally, the methods to investigate slip boundary conditions for nanoconfined liquids include theoretical analysis, physical experiments, and numerical simulations [8][24][25][26][27][28][29][30][31][32][33][34]. In recent years, machine learning methods have also been

Molecular assemblies on surfaces: towards physical and electronic decoupling of organic molecules

• Sabine Maier and
• Meike Stöhr

Beilstein J. Nanotechnol. 2021, 12, 950–956, doi:10.3762/bjnano.12.71

• partial exploration of the potential energy landscape due to the complexity of the system. Recently developed structure search methods that combine machine learning with density functional theory provide the possibility of reliable structure identification of non-planar molecules, as demonstrated for the

The role of convolutional neural networks in scanning probe microscopy: a review

• Ido Azuri,
• Irit Rosenhek-Goldian,
• Neta Regev-Rudzki,
• Georg Fantner and
• Sidney R. Cohen

Beilstein J. Nanotechnol. 2021, 12, 878–901, doi:10.3762/bjnano.12.66

• , various branches of machine learning have been the key facilitators in forging new paths, ranging from categorizing big data to instrumental control, from materials design through image analysis. Deep learning has the ability to identify abstract characteristics embedded within a data set, subsequently

• , convolutional neural networks, and how it is transforming the acquisition and analysis of scanning probe data. Keywords: atomic force microscopy (AFM); deep learning; machine learning; neural networks; scanning probe microscopy (SPM); Review Introduction: traditional machine learning vs deep learning Machine

• , regression (prediction of quantitative data values), translation (for instance of languages), anomaly detection, de-noising, clustering (grouping similar objects together), and data generation. In this review our major concern is with images, which are most relevant to certain aspects of machine learning as

Reducing molecular simulation time for AFM images based on super-resolution methods

• Zhipeng Dou,
• Jianqiang Qian,
• Yingzi Li,
• Rui Lin,
• Jianhai Wang,
• Peng Cheng and
• Zeyu Xu

Beilstein J. Nanotechnol. 2021, 12, 775–785, doi:10.3762/bjnano.12.61

• resolution. Theoretical investigations are getting highly important for the interpretation of AFM images. Researchers have used molecular simulation to examine the AFM imaging mechanism. With a recent flurry of researches applying machine learning to AFM, AFM images obtained from molecular simulation have

• can be used to speed up the generation of training data and vary simulation resolution for AFM machine learning. Keywords: atomic force microscopy; Bayesian compressed sensing; convolutional neural network; molecular dynamics simulation; super resolution; Introduction Atomic force microscopy methods

• , study the factors affecting resolution [16][17][18][19][20][21][22], and establish an appropriate simulation methodology for the explanation of complex imaging mechanism in liquids [23][24][25]. Besides, there has been a recent flurry of researches applying machine learning to AFM, including predicting

Nanogenerator-based self-powered sensors for data collection

• Yicheng Shao,
• Maoliang Shen,
• Yuankai Zhou,
• Xin Cui,
• Lijie Li and
• Yan Zhang

Beilstein J. Nanotechnol. 2021, 12, 680–693, doi:10.3762/bjnano.12.54

• used for gesture recognition [78][83][84]. The combination of a self-powered motion sensor and a back-end data processing system based on machine learning (ML) can realize sign language recognition for people with language impairment. Zhou et al. [84] fabricated a stretchable sensor for sign language

Intracranial recording in patients with aphasia using nanomaterial-based flexible electronics: promises and challenges

• Qingchun Wang and
• Wai Ting Siok

Beilstein J. Nanotechnol. 2021, 12, 330–342, doi:10.3762/bjnano.12.27

• using machine learning methods to examine the patients’ iEEG signals at the VWFA responding to words that differed in the degree of visual similarity, the researchers found that, shortly after stimulus onset (from approximately 100 to 430 ms), discriminating between words that shared no letters would

Detecting stable adsorbates of (1S)-camphor on Cu(111) with Bayesian optimization

• Jari Järvi,
• Patrick Rinke and
• Milica Todorović

Beilstein J. Nanotechnol. 2020, 11, 1577–1589, doi:10.3762/bjnano.11.140

• models, using, for example, a coarse-grained search space with discrete molecular configurations, or predetermined GP hyperparameters, at the cost of generality of the method. In this work, we show that the recently developed Bayesian Optimization Structure Search (BOSS) machine learning method [31][32

• , discuss our findings, and conclude the analysis. Computational Methods Adsorbate structure identification BOSS is a machine learning method that accelerates structure search via strategic sampling of the PES. With given initial data, BOSS builds the most probable surrogate model of the PES, refines it

• adsorbates, we concluded that in the most stable structures (class Ox), camphor chemisorbs to the Cu surface via O bonding. Our results imply that class Ox structures are viable candidates for static camphor adsorbates observed in AFM experiments. By combining machine learning with DFT, BOSS provides a novel

A novel method to remove impulse noise from atomic force microscopy images based on Bayesian compressed sensing

• Yingxu Zhang,
• Yingzi Li,
• Zihang Song,
• Zhenyu Wang,
• Jianqiang Qian and
• Junen Yao

Beilstein J. Nanotechnol. 2019, 10, 2346–2356, doi:10.3762/bjnano.10.225

• further improve the denoising performance, machine learning [10] and neural networks [11][12] are introduced to help remove the impulse noise. First, machine learning or neural networks are used to improve the accuracy of the recognition of noisy pixels. Then, the noise pixels are replaced by the median

• density. In addition, impulse noise filtering methods using machine learning [10], support vector machines [38], or neural networks [12] encounter the same problem as the adaptive median filter. When the noise density is lower than 0.5, the values of PSNR and SSIM acquired by the proposed method remain

Use of data processing for rapid detection of the prostate-specific antigen biomarker using immunomagnetic sandwich-type sensors

• Camila A. Proença,
• Tayane A. Freitas,
• Thaísa A. Baldo,
• Elsa M. Materón,
• Flávio M. Shimizu,
• Gabriella R. Ferreira,
• Frederico L. F. Soares,
• Ronaldo C. Faria and
• Osvaldo N. Oliveira Jr.

Beilstein J. Nanotechnol. 2019, 10, 2171–2181, doi:10.3762/bjnano.10.210

• cases with conjunction with machine learning approaches [31][59]. Here we employed multidimensional projection techniques based on linear and nonlinear multidimensional scaling (MDS) approaches such as principal component analysis (PCA) [60], Least squares projection (LSP) [61], Sammon’s mapping (SM

Review of advanced sensor devices employing nanoarchitectonics concepts

• Katsuhiko Ariga,
• Tatsuyuki Makita,
• Masato Ito,
• Taizo Mori,
• Shun Watanabe and
• Jun Takeya

Beilstein J. Nanotechnol. 2019, 10, 2014–2030, doi:10.3762/bjnano.10.198

• , and sampling of analytes. This roadmap would be shortened by using new types of materials such as two-dimensional materials [201][202][203][204] and through introducing new methodologies such as mass-data analyses and machine learning [205][206]. Another important factor to accelerate progress would

Adiabatic superconducting cells for ultra-low-power artificial neural networks

• Andrey E. Schegolev,
• Nikolay V. Klenov,
• Igor I. Soloviev and
• Maxim V. Tereshonok

Beilstein J. Nanotechnol. 2016, 7, 1397–1403, doi:10.3762/bjnano.7.130

• application in the fields of artificial intelligence and machine learning [1]. The future of cellular and satellite communications, radar systems, deep sea and space exploration will likely be closely related to the capability of ANNs to provide effective solutions to problems such as classification and

Nanoinformatics for environmental health and biomedicine

• Rong Liu and
• Yoram Cohen

Beilstein J. Nanotechnol. 2015, 6, 2449–2451, doi:10.3762/bjnano.6.253

• , literature mining for nano-data collection and meta-analysis, data mining/machine learning of nano-data (e.g., development of quantitative structure–activity relationships (QSARs)), simulation of the fate and transport of nanomaterials, nano-bio interactions, and assessment of potential environmental and

• for brain cancer [13]. As an imported aspect of nanoinformatics, recent advances in data mining/machine learning of nano-data are also reported in this Thematic Series. In one study, the toxicity of ZnO nanoparticles to zebrafish (measured by mortality rate (%)) was correlated to two principal

Predicting cytotoxicity of PAMAM dendrimers using molecular descriptors

• David E. Jones,
• Hamidreza Ghandehari and
• Julio C. Facelli

Beilstein J. Nanotechnol. 2015, 6, 1886–1896, doi:10.3762/bjnano.6.192

• journal articles. The results indicate that data mining and machine learning can be effectively used to predict the cytotoxicity of PAMAM dendrimers on Caco-2 cells. Keywords: data mining; machine learning; molecular descriptors; poly(amido amine) dendrimers (PAMAM); Introduction In silico approaches

• , such as data mining and machine learning, have been very successful in medicinal chemistry and are commonly used to guide the design of small pharmaceutical compounds [1]. In contrast, although nanomedicine is a rapidly growing field [2], there have been only a few attempts to use data mining

• viability of seven different cell lines [4]. Sayes and Ivanov used machine learning to predict the induced cellular membrane damage of immortalized human lung epithelial cells caused by metal oxide nanomaterials [5]. As discussed in a previous paper [6], there are a very limited number of databases

Framework for automatic information extraction from research papers on nanocrystal devices

• Thaer M. Dieb,
• Masaharu Yoshioka,
• Shinjiro Hara and
• Marcus C. Newton

Beilstein J. Nanotechnol. 2015, 6, 1872–1882, doi:10.3762/bjnano.6.190

• using the NaDev corpus and machine-learning techniques. However, the characteristics of NaDevEx were not examined in detail. In this paper, we conduct system evaluation experiments for NaDevEx using the NaDev corpus. We discuss three main issues: system performance, compared with human annotators; the

• Device Development corpus) [27][28] for research papers on nanocrystal device development. We also proposed a framework to extract information from research papers by using machine learning tools [29][30]. However, this system was only evaluated using the corpus constructed in our preliminary experiment

• , different approaches compete to extract chemical entities and drug names automatically from the literature [17] using the BioCreative IV CHEMDNER corpus [14]. We can classify approaches to information extraction and named entity recognition into two groups. One is a machine-learning approach that uses a

The eNanoMapper database for nanomaterial safety information

• Nina Jeliazkova,
• Charalampos Chomenidis,
• Philip Doganis,
• Bengt Fadeel,
• Roland Grafström,
• Barry Hardy,
• Janna Hastings,
• Markus Hegi,
• Vedrin Jeliazkov,
• Nikolay Kochev,
• Pekka Kohonen,
• Cristian R. Munteanu,
• Haralambos Sarimveis,
• Bart Smeets,
• Pantelis Sopasakis,
• Georgia Tsiliki,
• David Vorgrimmler and
• Egon Willighagen

Beilstein J. Nanotechnol. 2015, 6, 1609–1634, doi:10.3762/bjnano.6.165

• analyze it with multiple data preprocessing and machine learning algorithms. Conclusion: We demonstrate how the eNanoMapper database is used to import and publish online ENM and assay data from several data sources, how the “representational state transfer” (REST) API enables building user friendly

• . The red colour of the dots was chosen arbitrarily, but could reflect another feature, possibly the data sources as shown in the first example. Modelling The OpenTox API implementations contain all major statistical and machine learning (ML) algorithms required for the development of regression

• , data mining and machine learning algorithms and methods for defining the applicability domain of a predictive model. A screenshot of the Jaqpot web services is presented in Figure 19. Jaqpot provides asynchronous execution of tasks submitted by users, authentication, authorisation and accounting

Using natural language processing techniques to inform research on nanotechnology

• Nastassja A. Lewinski and
• Bridget T. McInnes

Beilstein J. Nanotechnol. 2015, 6, 1439–1449, doi:10.3762/bjnano.6.149

• entity extraction method using the supervised machine learning algorithm, support vector machines (SVM), was applied to their literature library. Supervised machine learning algorithms learn patterns and make predictions based on a set of training data. The training data for this system was generated by

• first parsing the text using a part-of-speech (POS) tagger, with tag category and boundary represented using the BIO format. The part-of-speech information, category, and context surrounding the term where used as features (or parameters) for the machine learning algorithm. For the source material, a

• CNER, renaming it SERB-CNER or syntactically enhanced rule-based chemical entity recognizer. SERB-CNER still focused on the Source Material tag. Here the POS tagger used was rb tagger. The machine learning system used was CRF++. This new system had recall improvements of 4–7% depending on which