RESUMO
Molecular dynamics (MD) simulations provide a powerful means of exploring the dynamic behavior of biomolecular systems at the atomic level. However, analyzing the vast data sets generated by MD simulations poses significant challenges. This article discusses the energy landscape visualization method (ELViM), a multidimensional reduction technique inspired by the energy landscape theory. ELViM transcends one-dimensional representations, offering a comprehensive analysis of the effective conformational phase space without the need for predefined reaction coordinates. We apply the ELViM to study the folding landscape of the antimicrobial peptide Polybia-MP1, showcasing its versatility in capturing complex biomolecular dynamics. Using dissimilarity matrices and a force-scheme approach, the ELViM provides intuitive visualizations, revealing structural correlations and local conformational signatures. The method is demonstrated to be adaptable, robust, and applicable to various biomolecular systems.
Assuntos
Simulação de Dinâmica Molecular , Termodinâmica , Conformação Proteica , Dobramento de Proteína , Peptídeos Antimicrobianos/químicaRESUMO
Molecular dynamics (MD) simulations have become increasingly powerful and can now describe the folding/unfolding of small biomolecules in atomic detail. However, a major challenge in MD simulations is to represent the complex energy landscape of biomolecules using a small number of reaction coordinates. In this study, we investigate the folding pathways of an RNA tetraloop, gcGCAAgc, using five classical MD simulations with a combined simulation time of approximately 120 µs. Our approach involves analyzing the tetraloop dynamics, including the folding transition state ensembles, using the energy landscape visualization method (ELViM). The ELViM is an approach that uses internal distances to compare any two conformations, allowing for a detailed description of the folding process without requiring root mean square alignment of structures. This method has previously been applied to describe the energy landscape of disordered ß-amyloid peptides and other proteins. The ELViM results in a non-linear projection of the multidimensional space, providing a comprehensive representation of the tetraloop's energy landscape. Our results reveal four distinct transition-state regions and establish the paths that lead to the folded tetraloop structure. This detailed analysis of the tetraloop's folding process has important implications for understanding RNA folding, and the ELViM approach can be used to study other biomolecules.
Assuntos
Peptídeos beta-Amiloides , Simulação de Dinâmica Molecular , RNARESUMO
Mastitis diagnosis can be made by detecting Staphylococcus aureus (S. aureus), which requires high sensitivity and selectivity. Here, we report on microfluidic genosensors and electronic tongues to detect S. aureus DNA using impedance spectroscopy with data analysis employing visual analytics and machine learning techniques. The genosensors were made with layer-by-layer films containing either 10 bilayers of chitosan/chondroitin sulfate or 8 bilayers of chitosan/sericin functionalized with an active layer of cpDNA S. aureus. The specific interactions leading to hybridization in these genosensors allowed for a low limit of detection of 5.90 × 10-19 mol/L. The electronic tongue had four sensing units made with 6-bilayer chitosan/chondroitin sulfate films, 10-bilayer chitosan/chondroitin sulfate, 8-bilayer chitosan/sericin, and 8-bilayer chitosan/gold nanoparticles modified with sericin. Despite the absence of specific interactions, various concentrations of DNA S. aureus could be distinguished when the impedance data were plotted using a dimensionality reduction technique. Selectivity of S. aureus DNA was confirmed using multidimensional calibration spaces, based on machine learning, with accuracy up to 89 % for the genosensors and 66 % for the electronic tongue. Hence, with these computational methods one may opt for the more expensive genosensors or the simpler and cheaper electronic tongue, depending on the sensitivity level required to diagnose mastitis.
Assuntos
Técnicas Biossensoriais , Quitosana , Staphylococcus aureus , Técnicas Biossensoriais/métodos , Calibragem , Quitosana/química , Espectroscopia Dielétrica/métodos , DNA Bacteriano/genética , DNA Bacteriano/análise , Nariz Eletrônico , Ouro/química , Staphylococcus aureus/isolamento & purificação , Staphylococcus aureus/genéticaRESUMO
The need to increase food production to address the world population growth can only be fulfilled with precision agriculture strategies to increase crop yield with minimal expansion of the cultivated area. One example is site-specific fertilization based on accurate monitoring of soil nutrient levels, which can be made more cost-effective using sensors. This study developed an impedimetric multisensor array using ion-selective membranes to analyze soil samples enriched with macronutrients (N, P, and K), which is compared with another array based on layer-by-layer films. The results obtained from both devices are analyzed with multidimensional projection techniques and machine learning methods, where a decision tree model algorithm chooses the calibrations (best frequencies and sensors). The multicalibration space method indicates that both devices effectively distinguished all soil samples tested, with the ion-selective membrane setup presenting a higher sensitivity to K content. These findings pave the way for more environmentally friendly and efficient agricultural practices, facilitating the mapping of cropping areas for precise fertilizer application and optimized crop yield.
RESUMO
The control of molecular architectures has been exploited in layer-by-layer (LbL) films deposited on Au interdigitated electrodes, thus forming an electronic tongue (e-tongue) system that reached an unprecedented high sensitivity (down to 10(-12) M) in detecting catechol. Such high sensitivity was made possible upon using units containing the enzyme tyrosinase, which interacted specifically with catechol, and by processing impedance spectroscopy data with information visualization methods. These latter methods, including the parallel coordinates technique, were also useful for identifying the major contributors to the high distinguishing ability toward catechol. Among several film architectures tested, the most efficient had a tyrosinase layer deposited atop LbL films of alternating layers of dioctadecyldimethylammonium bromide (DODAB) and 1,2-dipalmitoyl-sn-3-glycero-fosfo-rac-(1-glycerol) (DPPG), viz., (DODAB/DPPG)5/DODAB/Tyr. The latter represents a more suitable medium for immobilizing tyrosinase when compared to conventional polyelectrolytes. Furthermore, the distinction was more effective at low frequencies where double-layer effects on the film/liquid sample dominate the electrical response. Because the optimization of film architectures based on information visualization is completely generic, the approach presented here may be extended to designing architectures for other types of applications in addition to sensing and biosensing.
RESUMO
Dimensionality Reduction (DR) scatterplot layouts have become a ubiquitous visualization tool for analyzing multidimensional datasets. Despite their popularity, such scatterplots suffer from occlusion, especially when informative glyphs are used to represent data instances, potentially obfuscating critical information for the analysis under execution. Different strategies have been devised to address this issue, either producing overlap-free layouts that lack the powerful capabilities of contemporary DR techniques in uncovering interesting data patterns or eliminating overlaps as a post-processing strategy. Despite the good results of post-processing techniques, most of the best methods typically expand or distort the scatterplot area, thus reducing glyphs' size (sometimes) to unreadable dimensions, defeating the purpose of removing overlaps. This paper presents Distance Grid (DGrid), a novel post-processing strategy to remove overlaps from DR layouts that faithfully preserves the original layout's characteristics and bounds the minimum glyph sizes. We show that DGrid surpasses the state-of-the-art in overlap removal (through an extensive comparative evaluation considering multiple different metrics) while also being one of the fastest techniques, especially for large datasets. A user study with 51 participants also shows that DGrid is consistently ranked among the top techniques for preserving the original scatterplots' visual characteristics and the aesthetics of the final results.
RESUMO
Smartphone use provides a significant amount of screen-time for youth, and there have been growing concerns regarding its impact on their mental health. While time spent in a passive manner on the device is frequently considered deleterious, more active engagement with the phone might be protective for mental health. Recent developments in mobile sensing technology provide a unique opportunity to examine behaviour in a naturalistic manner. The present study sought to investigate, in a sample of 451 individuals (mean age 20.97 years old, 83% female), whether the amount of time spent on the device, an indicator of passive smartphone use, would be associated with worse mental health in youth and whether an active form of smartphone use, namely frequent checking of the device, would be associated with better outcomes. The findings highlight that overall time spent on the smartphone was associated with more pronounced internalizing and externalizing symptoms in youth, while the number of unlocks was associated with fewer internalizing symptoms. For externalizing symptoms, there was also a significant interaction between the two types of smartphone use observed. Using objective measures, our results suggest interventions targeting passive smartphone use may contribute to improving the mental health of youth.
Assuntos
COVID-19 , Aplicativos Móveis , Humanos , Feminino , Adolescente , Adulto Jovem , Adulto , Masculino , Smartphone , Saúde Mental , PandemiasRESUMO
The wide variety of molecular architectures used in sensors and biosensors and the large amount of data generated with some principles of detection have motivated the use of computational methods, such as information visualization techniques, not only to handle the data but also to optimize sensing performance. In this study, we combine projection techniques with micro-Raman scattering and atomic force microscopy (AFM) to address critical issues related to practical applications of electronic tongues (e-tongues) based on impedance spectroscopy. Experimentally, we used sensing units made with thin films of a perylene derivative (AzoPTCD acronym), coating Pt interdigitated electrodes, to detect CuCl(2) (Cu(2+)), methylene blue (MB), and saccharose in aqueous solutions, which were selected due to their distinct molecular sizes and ionic character in solution. The AzoPTCD films were deposited from monolayers to 120 nm via Langmuir-Blodgett (LB) and physical vapor deposition (PVD) techniques. Because the main aspects investigated were how the interdigitated electrodes are coated by thin films (architecture on e-tongue) and the film thickness, we decided to employ the same material for all sensing units. The capacitance data were projected into a 2D plot using the force scheme method, from which we could infer that at low analyte concentrations the electrical response of the units was determined by the film thickness. Concentrations at 10 µM or higher could be distinguished with thinner films--tens of nanometers at most--which could withstand the impedance measurements, and without causing significant changes in the Raman signal for the AzoPTCD film-forming molecules. The sensitivity to the analytes appears to be related to adsorption on the film surface, as inferred from Raman spectroscopy data using MB as analyte and from the multidimensional projections. The analysis of the results presented may serve as a new route to select materials and molecular architectures for novel sensors and biosensors, in addition to suggesting ways to unravel the mechanisms behind the high sensitivity obtained in various sensors.
Assuntos
Serviços de Informação , Perileno/análogos & derivados , Microscopia de Força Atômica , Perileno/químicaRESUMO
Multivariate or multidimensional visualization plays an essential role in exploratory data analysis by allowing users to derive insights and formulate hypotheses. Despite their popularity, it is usually users' responsibility to (visually) discover the data patterns, which can be cumbersome and time-consuming. Visual Analytics (VA) and machine learning techniques can be instrumental in mitigating this problem by automatically discovering and representing such patterns. One example is the integration of classification models with (visual) interpretability strategies, where models are used as surrogates for data patterns so that understanding a model enables understanding the phenomenon represented by the data. Although useful and inspiring, the few proposed solutions are based on visual representations of so-called black-box models, so the interpretation of the patterns captured by the models is not straightforward, requiring mechanisms to transform them into human-understandable pieces of information. This paper presents multiVariate dAta eXplanation (VAX), a new VA method to support identifying and visual interpreting patterns in multivariate datasets. Unlike the existing similar approaches, VAX uses the concept of Jumping Emerging Patterns, inherent interpretable logic statements representing class-variable relationships (patterns) derived from random Decision Trees. The potential of VAX is shown through use cases employing two real-world datasets covering different scenarios where intricate patterns are discovered and represented, something challenging to be done using usual exploratory approaches.
RESUMO
The diagnosis of cancer and other diseases using data from non-specific sensors - such as the electronic tongues (e-tongues) - is challenging owing to the lack of selectivity, in addition to the variability of biological samples. In this study, we demonstrate that impedance data obtained with an e-tongue in saliva samples can be used to diagnose cancer in the mouth. Data taken with a single-response microfluidic e-tongue applied to the saliva of 27 individuals were treated with multidimensional projection techniques and non-supervised and supervised machine learning algorithms. The distinction between healthy individuals and patients with cancer on the floor of mouth or oral cavity could only be made with supervised learning. Accuracy above 80% was obtained for the binary classification (YES or NO for cancer) using a Support Vector Machine (SVM) with radial basis function kernel and Random Forest. In the classification considering the type of cancer, the accuracy dropped to ca. 70%. The accuracy tended to increase when clinical information such as alcohol consumption was used in conjunction with the e-tongue data. With the random forest algorithm, the rules to explain the diagnosis could be identified using the concept of Multidimensional Calibration Space. Since the training of the machine learning algorithms is believed to be more efficient when the data of a larger number of patients are employed, the approach presented here is promising for computer-assisted diagnosis.
Assuntos
Neoplasias Bucais , Saliva , Algoritmos , Nariz Eletrônico , Humanos , Aprendizado de Máquina , Neoplasias Bucais/diagnóstico , Máquina de Vetores de SuporteRESUMO
Low-cost sensors to detect cancer biomarkers with high sensitivity and selectivity are essential for early diagnosis. Herein, an immunosensor was developed to detect the cancer biomarker p53 antigen in MCF7 lysates using electrical impedance spectroscopy. Interdigitated electrodes were screen printed on bacterial nanocellulose substrates, then coated with a matrix of layer-by-layer films of chitosan and chondroitin sulfate onto which a layer of anti-p53 antibodies was adsorbed. The immunosensing performance was optimized with a 3-bilayer matrix, with detection of p53 in MCF7 cell lysates at concentrations between 0.01 and 1000 Ucell. mL-1, and detection limit of 0.16 Ucell mL-1. The effective buildup of the immunosensor on bacterial nanocellulose was confirmed with polarization-modulated infrared reflection absorption spectroscopy (PM-IRRAS) and surface energy analysis. In spite of the high sensitivity, full selectivity with distinction of the p53-containing cell lysates and possible interferents required treating the data with a supervised machine learning approach based on decision trees. This allowed the creation of a multidimensional calibration space with 11 dimensions (frequencies used to generate decision tree rules), with which the classification of the p53-containing samples can be explained.
Assuntos
Técnicas Biossensoriais , Neoplasias , Biomarcadores Tumorais/análise , Espectroscopia Dielétrica , Eletrodos , ImunoensaioRESUMO
The development of new methods and concepts to visualize massive amounts of data holds the promise to revolutionize the way scientific results are analyzed, especially when tasks such as classification and clustering are involved, as in the case of sensing and biosensing. In this paper we employ a suite of software tools, referred to as PEx-Sensors, through which projection techniques are used to analyze electrical impedance spectroscopy data in electronic tongues and related sensors. The possibility of treating high dimension datasets with PEx-Sensors is advantageous because the whole impedance vs. frequency curves obtained with various sensing units and for a variety of samples can be analyzed at once. It will be shown that non-linear projection techniques such as Sammon's Mapping or IDMAP provide higher distinction ability than linear methods for sensor arrays containing units capable of molecular recognition, apparently because these techniques are able to capture the cooperative response owing to specific interactions between the sensing unit material and the analyte. In addition to allowing for a higher sensitivity and selectivity, the use of PEx-Sensors permits the identification of the major contributors for the distinguishing ability of sensing units and of the optimized frequency range. The latter will be illustrated with sensing units made with layer-by-layer (LbL) films to detect phytic acid, whose capacitance data were visualized with Parallel Coordinates. Significantly, the implementation of PEx-Sensors was conceived so as to handle any type of sensor based on any type of principle of detection, representing therefore a generic platform for treating large amounts of data for sensors and biosensors.
Assuntos
Técnicas Biossensoriais/métodos , Técnicas Biossensoriais/instrumentação , Espectroscopia Dielétrica/métodos , Eletrônica , SoftwareRESUMO
Recent advances in the control of molecular engineering architectures have allowed unprecedented ability of molecular recognition in biosensing, with a promising impact for clinical diagnosis and environment control. The availability of large amounts of data from electrical, optical, or electrochemical measurements requires, however, sophisticated data treatment in order to optimize sensing performance. In this study, we show how an information visualization system based on projections, referred to as Projection Explorer (PEx), can be used to achieve high performance for biosensors made with nanostructured films containing immobilized antigens. As a proof of concept, various visualizations were obtained with impedance spectroscopy data from an array of sensors whose electrical response could be specific toward a given antibody (analyte) owing to molecular recognition processes. In addition to discussing the distinct methods for projection and normalization of the data, we demonstrate that an excellent distinction can be made between real samples tested positive for Chagas disease and Leishmaniasis, which could not be achieved with conventional statistical methods. Such high performance probably arose from the possibility of treating the data in the whole frequency range. Through a systematic analysis, it was inferred that Sammon's mapping with standardization to normalize the data gives the best results, where distinction could be made of blood serum samples containing 10(-7) mg/mL of the antibody. The method inherent in PEx and the procedures for analyzing the impedance data are entirely generic and can be extended to optimize any type of sensor or biosensor.
Assuntos
Técnicas Biossensoriais/métodos , Leishmaniose/diagnóstico , Anticorpos Antiprotozoários/sangue , Antígenos de Protozoários , Doença de Chagas/diagnóstico , Impedância Elétrica , NanoestruturasRESUMO
Over the past decades, classification models have proven to be essential machine learning tools given their potential and applicability in various domains. In these years, the north of the majority of the researchers had been to improve quantitative metrics, notwithstanding the lack of information about models' decisions such metrics convey. This paradigm has recently shifted, and strategies beyond tables and numbers to assist in interpreting models' decisions are increasing in importance. Part of this trend, visualization techniques have been extensively used to support classification models' interpretability, with a significant focus on rule-based models. Despite the advances, the existing approaches present limitations in terms of visual scalability, and the visualization of large and complex models, such as the ones produced by the Random Forest (RF) technique, remains a challenge. In this paper, we propose Explainable Matrix (ExMatrix), a novel visualization method for RF interpretability that can handle models with massive quantities of rules. It employs a simple yet powerful matrix-like visual metaphor, where rows are rules, columns are features, and cells are rules predicates, enabling the analysis of entire models and auditing classification results. ExMatrix applicability is confirmed via different examples, showing how it can be used in practice to promote RF models interpretability.
RESUMO
The integration of nanostructured films containing biomolecules and silicon-based technologies is a promising direction for reaching miniaturized biosensors that exhibit high sensitivity and selectivity. A challenge, however, is to avoid cross talk among sensing units in an array with multiple sensors located on a small area. In this letter, we describe an array of 16 sensing units of a light-addressable potentiometric sensor (LAPS), which was made with layer-by-layer (LbL) films of a poly(amidomine) dendrimer (PAMAM) and single-walled carbon nanotubes (SWNTs), coated with a layer of the enzyme penicillinase. A visual inspection of the data from constant-current measurements with liquid samples containing distinct concentrations of penicillin, glucose, or a buffer indicated a possible cross talk between units that contained penicillinase and those that did not. With the use of multidimensional data projection techniques, normally employed in information visualization methods, we managed to distinguish the results from the modified LAPS, even in cases where the units were adjacent to each other. Furthermore, the plots generated with the interactive document map (IDMAP) projection technique enabled the distinction of the different concentrations of penicillin, from 5 mmol L(-1) down to 0.5 mmol L(-1). Data visualization also confirmed the enhanced performance of the sensing units containing carbon nanotubes, consistent with the analysis of results for LAPS sensors. The use of visual analytics, as with projection methods, may be essential to handle a large amount of data generated in multiple sensor arrays to achieve high performance in miniaturized systems.
Assuntos
Técnicas Biossensoriais/instrumentação , Potenciometria/instrumentação , Potenciometria/métodos , Glucose/química , Luz , Nanotubos de Carbono , Penicilina G/química , Sensibilidade e EspecificidadeRESUMO
Impedance spectroscopy has been proven a powerful tool for reaching high sensitivity in sensor arrays made with nanostructured films in the so-called electronic tongue systems, whose distinguishing ability may be enhanced with sensing units capable of molecular recognition. In this study we show that for optimized sensors and biosensors the dielectric relaxation processes involved in impedance measurements should also be considered, in addition to an adequate choice of sensing materials. We used sensing units made from layer-by-layer (LbL) films with alternating layers of the polyeletrolytes, poly(allylamine) hydrochloride (PAH) and poly(vinyl sulfonate) (PVS), or LbL films of PAH alternated with layers of the enzyme phytase, all adsorbed on gold interdigitate electrodes. Surprisingly, the detection of phytic acid was as effective in the PVS/PAH sensing system as with the PAH/phytase system, in spite of the specific interactions of the latter. This was attributed to the dependence of the relaxation processes on nonspecific interactions such as electrostatic cross-linking and possibly on the distinct film architecture as the phytase layers were found to grow as columns on the LbL film, in contrast to the molecularly thin PAH/PVS films. Using projection techniques, we were able to detect phytic acid at the micromolar level with either of the sensing units in a data analysis procedure that allows for further optimization.
Assuntos
Técnicas Biossensoriais/métodos , Ácido Fítico/química , Impedância Elétrica , Eletrodos , Ouro/química , Ácido Fítico/análise , Poliaminas/química , Polivinil/química , Ácidos Sulfônicos/químicaRESUMO
The need for reliable, fast diagnostics is closely linked to the need for safe, effective treatment of the so-called "neglected" diseases. The list of diseases with no field-adapted diagnostic tools includes leishmaniasis, shigella, typhoid, and bacterial meningitis. Leishmaniasis, in particular, is a parasitic disease caused by Leishmania spp. transmitted by infected phlebotomine sandfly, which remains a public health concern in developing countries with ca. 12 million people infected and 350 million at risk of infection. Despite several attempts, methods for diagnosis are still noneffective, especially with regard to specificity due to false positives with Chagas' disease caused by Trypanosoma cruzi . Accepted golden standards for detecting leishmaniasis involve isolation of parasites either microscopically, or by culture, and in both methods specimens are obtained by invasive means. Here, we show that efficient distinction between cutaneous leishmaniasis and Chagas' disease can be obtained with a low-cost biosensor system made with nanostructured films containing specific Leishmania amazonensis and T. cruzi antigens and employing impedance spectroscopy as the detection method. This unprecedented selectivity was afforded by antigen-antibody molecular recognition processes inherent in the detection with the immobilized antigens, and by statistically correlating the electrical impedance data, which allowed distinction between real samples that tested positive for Chagas' disease and leishmaniasis. Distinction could be made of blood serum samples containing 10(-5) mg/mL of the antibody solution in a few minutes. The methods used here are generic and can be extended to any type of biosensor, which is important for an effective diagnosis of many other diseases.
Assuntos
Técnicas Biossensoriais/métodos , Leishmaniose/diagnóstico , Animais , Anticorpos/sangue , Antígenos/química , Antígenos/imunologia , Doença de Chagas/diagnóstico , Dendrímeros/química , Técnicas Eletroquímicas/métodos , Eletrodos , Proteínas Imobilizadas/química , Proteínas Imobilizadas/imunologia , Leishmania/imunologia , Camundongos , Nanoestruturas/química , Doenças Negligenciadas/diagnóstico , Análise de Componente Principal , Trypanosoma cruzi/imunologiaRESUMO
Most multidimensional projection techniques rely on distance (dissimilarity) information between data instances to embed high-dimensional data into a visual space. When data are endowed with Cartesian coordinates, an extra computational effort is necessary to compute the needed distances, making multidimensional projection prohibitive in applications dealing with interactivity and massive data. The novel multidimensional projection technique proposed in this work, called Part-Linear Multidimensional Projection (PLMP), has been tailored to handle multivariate data represented in Cartesian high-dimensional spaces, requiring only distance information between pairs of representative samples. This characteristic renders PLMP faster than previous methods when processing large data sets while still being competitive in terms of precision. Moreover, knowing the range of variation for data instances in the high-dimensional space, we can make PLMP a truly streaming data projection technique, a trait absent in previous methods.
RESUMO
Point placement strategies aim at mapping data points represented in higher dimensions to bi-dimensional spaces and are frequently used to visualize relationships amongst data instances.They have been valuable tools for analysis and exploration of datasets of various kinds. Many conventional techniques, however, do not behave well when the number of dimensions is high, such as in the case of documents collections. Later approaches handle that shortcoming, but may cause too much clutter to allow flexible exploration to take place. In this work we present a novel hierarchical point placement technique that is capable of dealing with these problems. While good grouping and separation of data with high similarity is maintained without increasing computation cost,its hierarchical structure lends itself both to exploration in various levels of detail and to handling data in subsets, improving analysis capability and also allowing manipulation of larger data sets.
RESUMO
The problem of projecting multidimensional data into lower dimensions has been pursued by many researchers due to its potential application to data analysis of various kinds. This paper presents a novel multidimensional projection technique based on least square approximations. The approximations compute the coordinates of a set of projected points based on the coordinates of a reduced number of control points with defined geometry. We name the technique Least Square Projections (LSP). From an initial projection of the control points, LSP defines the positioning of their neighboring points through a numerical solution that aims at preserving a similarity relationship between the points given by a metric in mD. In order to perform the projection, a small number of distance calculations is necessary and no repositioning of the points is required to obtain a final solution with satisfactory precision. The results show the capability of the technique to form groups of points by degree of similarity in 2D. We illustrate that capability through its application to mapping collections of textual documents from varied sources, a strategic yet difficult application. LSP is faster and more accurate than other existing high quality methods, particularly where it was mostly tested, that is, for mapping text sets.