The rise of deep learning and transformations in bioactivity prediction power of molecular modeling tools

2021 ◽  
Author(s):  
Mohammed Bule ◽  
Nafiseh Jalalimanesh ◽  
Zahra Bayrami ◽  
Maryam Baeeri ◽  
Mohammad Abdollahi
Keyword(s):  
Deep Learning ◽  
Molecular Modeling ◽  
Modeling Tools ◽  
Bioactivity Prediction

scholarly journals A Perspective on Deep Learning for Molecular Modeling and Simulations

2020 ◽  
Author(s):  
Jun Zhang ◽  
Yao-Kun Lei ◽  
Zhen Zhang ◽  
Junhan Chang ◽  
Maodong Li ◽  
...  
Keyword(s):  
Deep Learning ◽  
Molecular Modeling ◽  
Modeling And Simulations

scholarly journals Validating the validation: reanalyzing a large-scale comparison of deep learning and machine learning models for bioactivity prediction

2020 ◽  
Vol 34 (7) ◽  
pp. 717-730 ◽  
Author(s):  
Matthew C. Robinson ◽  
Robert C. Glen ◽  
Alpha A. Lee
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Numerical Experiments ◽  
Large Scale ◽  
Operating Characteristic ◽  
Characteristic Curve ◽  
Learning Models ◽  
Bioactivity Prediction ◽  
Operating Characteristic Curve ◽  
Machine Learning Models

Abstract Machine learning methods may have the potential to significantly accelerate drug discovery. However, the increasing rate of new methodological approaches being published in the literature raises the fundamental question of how models should be benchmarked and validated. We reanalyze the data generated by a recently published large-scale comparison of machine learning models for bioactivity prediction and arrive at a somewhat different conclusion. We show that the performance of support vector machines is competitive with that of deep learning methods. Additionally, using a series of numerical experiments, we question the relevance of area under the receiver operating characteristic curve as a metric in virtual screening. We further suggest that area under the precision–recall curve should be used in conjunction with the receiver operating characteristic curve. Our numerical experiments also highlight challenges in estimating the uncertainty in model performance via scaffold-split nested cross validation.

Deep learning for variational multiscale molecular modeling

2020 ◽  
Vol 153 (17) ◽  
pp. 174115
Author(s):  
Jun Zhang ◽  
Yao-Kun Lei ◽  
Yi Isaac Yang ◽  
Yi Qin Gao
Keyword(s):  
Deep Learning ◽  
Molecular Modeling ◽  
Variational Multiscale

scholarly journals Rational Design of Potential Bcr-Abl Tyrosine Kinase Inhibitors by the Methods of Molecular Modeling

2020 ◽  
Vol 15 (2) ◽  
pp. 396-415
Author(s):  
A.M. Anrdrianov ◽  
Yu.V. Kornoushenko ◽  
A.D. Karpenko ◽  
I.P. Bosko ◽  
Zh.V. Ignatovich ◽  
...  
Keyword(s):  
Molecular Modeling ◽  
Tyrosine Kinase ◽  
Rational Design ◽  
Kinase Inhibitors ◽  
Calculated Data ◽  
Protein Kinase Inhibitors ◽  
Modeling Tools ◽  
Abl Tyrosine Kinase ◽  
Anti Cancer ◽  
Derivatives Of

Discovery of the nature of inhibiting cancer processes by small organic molecules has changed the principles of the development of drug compounds for antitumor therapy. Recent achievements in this area are associated with the design of small-molecule protein kinase inhibitors, organic compounds exhibiting directed pathogenetic action. In this study, in silico design of 38 potential anti-cancer compounds with multikinase profile was carried out based on the derivatives of 2-arylaminopyrimidine. Evaluation of inhibitory activity potential of these compounds against the native and mutant (T315I) forms of Bcr-Abl tyrosine kinase, an enzyme that plays a key role in the pathogenesis of chronic myeloid leukemia characterized by uncontrolled growth myeloid cells in peripheral blood and bone marrow, was performed using molecular modeling tools. As a result, 5 top-ranking compounds that exhibit, according to the calculated data, a high-affinity binding to the native and mutant Bcr-Abl tyrosine kinase were identified. The designed compounds were shown to form good scaffolds for the development of novel potent antitumor drugs.

scholarly journals Automatic Crack Classification by Exploiting Statistical Event Descriptors for Deep Learning

2021 ◽  
Vol 11 (24) ◽  
pp. 12059
Author(s):  
Giulio Siracusano ◽  
Francesca Garescì ◽  
Giovanni Finocchio ◽  
Riccardo Tomasello ◽  
Francesco Lamonaca ◽  
...  
Keyword(s):  
Deep Learning ◽  
Short Term Memory ◽  
Low Cost ◽  
Acoustic Emissions ◽  
Ultrasonic Waves ◽  
Modeling Tools ◽  
Spectral Kurtosis ◽  
Long Short Term Memory ◽  
Classification Tool ◽  
Modern Building

In modern building infrastructures, the chance to devise adaptive and unsupervised data-driven structural health monitoring (SHM) systems is gaining in popularity. This is due to the large availability of big data from low-cost sensors with communication capabilities and advanced modeling tools such as deep learning. A promising method suitable for smart SHM is the analysis of acoustic emissions (AEs), i.e., ultrasonic waves generated by internal ruptures of the concrete when it is stressed. The advantage in respect to traditional ultrasonic measurement methods is the absence of the emitter and the suitability to implement continuous monitoring. The main purpose of this paper is to combine deep neural networks with bidirectional long short term memory and advanced statistical analysis involving instantaneous frequency and spectral kurtosis to develop an accurate classification tool for tensile, shear and mixed modes originated from AE events (cracks). We investigated effective event descriptors to capture the unique characteristics from the different types of modes. Tests on experimental results confirm that this method achieves promising classification among different crack events and can impact on the design of the future of SHM technologies. This approach is effective to classify incipient damages with 92% of accuracy, which is advantageous to plan maintenance.

Sign in / Sign up

Export Citation Format

Share Document