scholarly journals Molecular Design Using Signal Processing and Machine Learning: Time-Frequency-like Representation and Forward Design

2021 ◽  
Author(s):  
Alain Beaudelaire Tchagang ◽  
Ahmed H. Tewfik ◽  
Julio J. Valdés
Keyword(s):  
Machine Learning ◽  
Signal Processing ◽  
High Speed ◽  
Density Functional ◽  
Molecular Design ◽  
Absolute Error ◽  
Molecular Data ◽  
Deep Convolutional Neural Networks ◽  
Time Frequency ◽  
Short Time

Abstract Accumulation of molecular data obtained from quantum mechanics (QM) theories such as density functional theory (DFTQM) make it possible for machine learning (ML) to accelerate the discovery of new molecules, drugs, and materials. Models that combine QM with ML (QM↔ML) have been very effective in delivering the precision of QM at the high speed of ML. In this study, we show that by integrating well-known signal processing (SP) techniques (i.e. short time Fourier transform, continuous wavelet analysis and Wigner-Ville distribution) in the QM↔ML pipeline, we obtain a powerful machinery (QM↔SP↔ML) that can be used for representation, visualization and forward design of molecules. More precisely, in this study, we show that the time-frequency-like representation of molecules encodes their structural, geometric, energetic, electronic and thermodynamic properties. This is demonstrated by using the new representation in the forward design loop as input to a deep convolutional neural networks trained on DFTQM calculations, which outputs the properties of the molecules. Tested on the QM9 dataset (composed of 133,855 molecules and 16 properties), the new QM↔SP↔ML model is able to predict the properties of molecules with a mean absolute error (MAE) below acceptable chemical accuracy (i.e. MAE < 1 Kcal/mol for total energies and MAE < 0.1 ev for orbital energies). Furthermore, the new approach performs similarly or better compared to other ML state-of-the-art techniques described in the literature. In all, in this study, we show that the new QM↔SP↔ML model represents a powerful technique for molecular forward design. All the codes and data generated and used in this study are available as supporting materials. The QM↔SP↔ML is also housed at the following website: https://github.com/TABeau/QM-SP-ML.

Evaluation of potential auras in generalized epilepsy from EEG signals using deep convolutional neural networks and time-frequency representation

2020 ◽  
Vol 65 (4) ◽  
pp. 379-391 ◽  
Author(s):  
Hasan Polat ◽  
Mehmet Ufuk Aluçlu ◽  
Mehmet Siraç Özerdem
Keyword(s):  
Eeg Signals ◽  
Deep Convolutional Neural Networks ◽  
One Dimensional ◽  
Time Frequency ◽  
Specificity And Sensitivity ◽  
Frequency Representation ◽  
Proposed Model ◽  
Generalized Epilepsy ◽  
Short Time

AbstractThe general uncertainty of epilepsy and its unpredictable seizures often affect badly the quality of life of people exposed to this disease. There are patients who can be considered fortunate in terms of prediction of any seizures. These are patients with epileptic auras. In this study, it was aimed to evaluate pre-seizure warning symptoms of the electroencephalography (EEG) signals by a convolutional neural network (CNN) inspired by the epileptic auras defined in the medical field. In this context, one-dimensional EEG signals were transformed into a spectrogram display form in the frequency-time domain by applying a short-time Fourier transform (STFT). Systemic changes in pre-epileptic seizure have been described by applying the CNN approach to the EEG signals represented in the image form, and the subjective EEG-Aura process has been tried to be determined for each patient. Considering all patients included in the evaluation, it was determined that the 1-min interval covering the time from the second minute to the third minute before the seizure had the highest mean and the lowest variance to determine the systematic changes before the seizure. Thus, the highest performing process is described as EEG-Aura. The average success for the EEG-Aura process was 90.38 ± 6.28%, 89.78 ± 8.34% and 90.47 ± 5.95% for accuracy, specificity and sensitivity, respectively. Through the proposed model, epilepsy patients who do not respond to medical treatment methods are expected to maintain their lives in a more comfortable and integrated way.

scholarly journals Identification of Grain Oriented SiFe Steels Based on Imaging the Instantaneous Dynamics of Magnetic Barkhausen Noise Using Short-Time Fourier Transform and Deep Convolutional Neural Network

Materials ◽  
2021 ◽  
Vol 15 (1) ◽  
pp. 118
Author(s):  
Michal Maciusowicz ◽  
Grzegorz Psuj ◽  
Paweł Kochmański
Keyword(s):  
Fourier Transform ◽  
Barkhausen Noise ◽  
Magnetic Barkhausen Noise ◽  
Short Time Fourier Transform ◽  
Deep Convolutional Neural Networks ◽  
Electrical Steels ◽  
New Approach ◽  
Time Frequency ◽  
Short Time

This paper presents a new approach to the extraction and analysis of information contained in magnetic Barkhausen noise (MBN) for evaluation of grain oriented (GO) electrical steels. The proposed methodology for MBN analysis is based on the combination of the Short-Time Fourier Transform for the observation of the instantaneous dynamics of the phenomenon and deep convolutional neural networks (DCNN) for the extraction of hidden information and building the knowledge. The use of DCNN makes it possible to find even complex and convoluted rules of the Barkhausen phenomenon course, difficult to determine based solely on the selected features of MBN signals. During the tests, several samples made of conventional and high permeability GO steels were tested at different angles between the rolling and transverse directions. The influences of the angular resolution and the proposed additional prediction update algorithm on the DCNN accuracy were investigated, obtaining the highest gain for the angle of 3.6°, for which the overall accuracy exceeded 80%. The obtained results indicate that the proposed new solution combining time–frequency analysis and DCNN for the quantification of information from MBN having stochastic nature may be a very effective tool in the characterization of the magnetic materials.

scholarly journals Looking Beyond Adsorption Energies to Understand Interactions at Surface Using Machine Learning

2021 ◽  
Author(s):  
Sheena Agarwal ◽  
Kavita Joshi
Keyword(s):  
Machine Learning ◽  
Small Molecules ◽  
Density Functional ◽  
Mean Absolute Error ◽  
Absolute Error ◽  
Face Centered Cubic ◽  
Functional Theory ◽  
Full Picture ◽  
Adsorption Energies ◽  
Face Centered

Abstract<br>Identifying factors that influence interactions at the surface is still an active area of research. In this study, we present the importance of analyzing bondlength activation, while interpreting Density Functional Theory (DFT) results, as yet another crucial indicator for catalytic activity. We studied the<br>adsorption of small molecules, such as O 2 , N 2 , CO, and CO 2 , on seven face-centered cubic (fcc) transition metal surfaces (M = Ag, Au, Cu, Ir, Rh, Pt, and Pd) and their commonly studied facets (100, 110, and 111). Through our DFT investigations, we highlight the absence of linear correlation between adsorption energies (E ads ) and bondlength activation (BL act ). Our study indicates the importance of evaluating both to develop a better understanding of adsorption at surfaces. We also developed a Machine Learning (ML) model trained on simple periodic table properties to predict both, E ads and BL act . Our ML model gives an accuracy of Mean Absolute Error (MAE) ∼ 0.2 eV for E ads predictions and 0.02 Å for BL act predictions. The systematic study of the ML features<br>that affect E ads and BL act further reinforces the importance of looking beyond adsorption energies to get a full picture of surface interactions with DFT.<br>

Supervised Machine Learning Algorithms for Evaluation of Solid Lipid Nanoparticles and Particle Size

2019 ◽  
Vol 21 (9) ◽  
pp. 693-699 ◽  
Author(s):  
A. Alper Öztürk ◽  
A. Bilge Gündüz ◽  
Ozan Ozisik
Keyword(s):  
Machine Learning ◽  
Particle Size ◽  
High Speed ◽  
Mean Absolute Error ◽  
Mixing Time ◽  
Learning Algorithms ◽  
Absolute Error ◽  
Machine Learning Algorithms ◽  
Learning Methods ◽  
Machine Learning Methods

Aims and Objectives: Solid Lipid Nanoparticles (SLNs) are pharmaceutical delivery systems that have advantages such as controlled drug release, long-term stability etc. Particle Size (PS) is one of the important criteria of SLNs. These factors affect drug release rate, bio-distribution etc. In this study, the formulation of SLNs using high-speed homogenization technique has been evaluated. The main emphasis of the work is to study whether the effect of mixing time and formulation ingredients on PS can be modeled. For this purpose, different machine learning algorithms have been applied and evaluated using the mean absolute error metric. Materials and Methods: SLNs were prepared by high-speed homogenizaton. PS, size distribution and zeta potential measurements were performed on freshly prepared samples. In order to model the formulation of the particles in terms of mixing time and formulation ingredients and evaluate the predictability of PS depending on these parameters, different machine learning algorithms were applied on the prepared dataset and the performances of the algorithms were also evaluated. Results: PS of SLNs obtained was in the range of 263-498nm. The results present that PS of SLNs can be best estimated by decision tree based methods, among which Random Forest has the least mean absolute error value with 0.028. As a result, the estimation of machine learning algorithms demonstrates that particle size can be estimated by both decision rule-based machine learning methods and function fitting machine learning methods. Conclusion: Our findings present that machine learning methods can be highly useful for determining formulation parameters for further research.

scholarly journals Machine learning based prediction of lattice thermal conductivity for half-Heusler compounds using atomic information

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Hidetoshi Miyazaki ◽  
Tomoyuki Tamura ◽  
Masashi Mikami ◽  
Kosuke Watanabe ◽  
Naoki Ide ◽  
...  
Keyword(s):  
Machine Learning ◽  
Thermal Conductivity ◽  
Density Functional ◽  
Lattice Thermal Conductivity ◽  
Low Cost ◽  
Functional Materials ◽  
Atomic Radius ◽  
Heusler Compounds ◽  
Heusler Compound ◽  
Short Time

AbstractHalf-Heusler compound has drawn attention in a variety of fields as a candidate material for thermoelectric energy conversion and spintronics technology. When the half-Heusler compound is incorporated into the device, the control of high lattice thermal conductivity owing to high crystal symmetry is a challenge for the thermal manager of the device. The calculation for the prediction of lattice thermal conductivity is an important physical parameter for controlling the thermal management of the device. We examined whether lattice thermal conductivity prediction by machine learning was possible on the basis of only the atomic information of constituent elements for thermal conductivity calculated by the density functional theory in various half-Heusler compounds. Consequently, we constructed a machine learning model, which can predict the lattice thermal conductivity with high accuracy from the information of only atomic radius and atomic mass of each site in the half-Heusler type crystal structure. Applying our results, the lattice thermal conductivity for an unknown half-Heusler compound can be immediately predicted. In the future, low-cost and short-time development of new functional materials can be realized, leading to breakthroughs in the search of novel functional materials.

Δ-Quantum machine learning for medicinal chemistry

2021 ◽  
Author(s):  
Kenneth Atz ◽  
Clemens Isert ◽  
Markus N. A. Böcker ◽  
José Jiménez-Luna ◽  
Gisbert Schneider
Keyword(s):  
Machine Learning ◽  
Message Passing ◽  
Density Functional ◽  
Large Scale ◽  
Molecular Design ◽  
Low Cost ◽  
Computational Cost ◽  
Three Dimensional ◽  
Biomolecular Systems ◽  
Quantum Observables

Many molecular design tasks benefit from fast and accurate calculations of quantum-mechanical (QM) properties. However, the computational cost of QM methods applied to drug-like molecules currently renders large-scale applications of quantum chemistry challenging. Aiming to mitigate this problem, we developed DelFTa, an open-source toolbox for the prediction of electronic properties of drug-like molecules at the density functional (DFT) level of theory, using Δ-machine-learning. Δ-Learning corrects the prediction error (Δ) of a fast but inaccurate property calculation. DelFTa employs state-of-the-art three-dimensional message-passing neural networks trained on a large dataset of QM properties. It provides access to a wide array of quantum observables on the molecular, atomic and bond levels by predicting approximations to DFT values from a low-cost semiempirical baseline. Δ-Learning outperformed its direct-learning counterpart for most of the considered QM endpoints. The results suggest that predictions for non-covalent intra- and intermolecular interactions can be extrapolated to larger biomolecular systems. The software is fully open-sourced and features documented command-line and Python APIs.

Ultra-High-Speed Time-Frequency Signal Processing

2020 ◽  
Vol 31 (12) ◽  
pp. 37
Author(s):  
Saikrishna Reddy Konatham ◽  
Reza Maram ◽  
Luis Romero Cortés ◽  
José Azaña ◽  
Jun Ho Chang ◽  
...  
Keyword(s):  
Signal Processing ◽  
High Speed ◽  
Frequency Signal ◽  
Time Frequency ◽  
Ultra High Speed

scholarly journals Time-frequency tools of signal processing for EISCAT data analysis

1996 ◽  
Vol 14 (12) ◽  
pp. 1513-1525
Author(s):  
J. Lilensten ◽  
P. O. Amblard
Keyword(s):  
Signal Processing ◽  
Fourier Transform ◽  
Data Analysis ◽  
Gravity Waves ◽  
Velocity Measurements ◽  
Time Frequency ◽  
Synthetic Signal ◽  
Stationary Analysis ◽  
The Fourier Transform ◽  
Short Time

Abstract. We demonstrate the usefulness of some signal-processing tools for the EISCAT data analysis. These tools are somewhat less classical than the familiar periodogram, squared modulus of the Fourier transform, and therefore not as commonly used in our community. The first is a stationary analysis, "Thomson's estimate'' of the power spectrum. The other two belong to time-frequency analysis: the short-time Fourier transform with the spectrogram, and the wavelet analysis via the scalogram. Because of the highly non-stationary character of our geophysical signals, the latter two tools are better suited for this analysis. Their results are compared with both a synthetic signal and EISCAT ion-velocity measurements. We show that they help to discriminate patterns such as gravity waves from noise.

scholarly journals Open-source Δ-quantum machine learning for medicinal chemistry

2021 ◽  
Author(s):  
Kenneth Atz ◽  
Clemens Isert ◽  
Markus N. A. Böcker ◽  
José Jiménez-Luna ◽  
Gisbert Schneider
Keyword(s):  
Machine Learning ◽  
Open Source ◽  
Density Functional ◽  
Large Scale ◽  
Molecular Design ◽  
State Of The Art ◽  
Computational Cost ◽  
Quantum Mechanical ◽  
Quantum Observables ◽  
Graph Neural Networks

Certain molecular design tasks benefit from fast and accurate calculations of quantum-mechanical (QM) properties. However, the computational cost of QM methods applied to drug-like compounds currently makes large-scale applications of quantum chemistry challenging. In order to mitigate this problem, we developed DelFTa, an open-source toolbox for predicting small-molecule electronic properties at the density functional (DFT) level of theory, using the Δ-machine learning principle. DelFTa employs state-of-the-art E(3)-equivariant graph neural networks that were trained on the QMugs dataset of QM properties. It provides access to a wide array of quantum observables by predicting approximations to ωB97X-D/def2-SVP values from a GFN2-xTB semiempirical baseline. Δ-learning with DelFTa was shown to outperform direct DFT learning for most of the considered QM endpoints. The software is provided as open-source code with fully-documented command-line and Python APIs.

Sign in / Sign up

Export Citation Format

Share Document