molecules-logo

Journal Browser

Journal Browser

Special Issue "Deep Learning for Molecular Structure Modelling"

A special issue of Molecules (ISSN 1420-3049). This special issue belongs to the section "Computational and Theoretical Chemistry".

Deadline for manuscript submissions: 15 September 2021.

Special Issue Editor

Dr. Liang Zhao
E-Mail Website
Guest Editor
George Mason University, Fairfax Campus, Fairfax, United States
Interests: Data mining and machine learning, especially on spatial, temporal, and network modeling and knolwedge discovery; deep learning on graphs; interpretable machine learning; gradient-free optimization for deep neural network training

Special Issue Information

Dear Colleagues,

The molecular structure is critical in determining chemical, physical, and biological properties. Despite significant research in molecular structure modeling over the last few decades, many outstanding challenges remain. From a computational perspective, molecules pose high-dimensional structural data that involve spatial, network, geometric, geodesic, and temporal characteristics which jointly determine molecular functions and properties. Grand challenges include how to concisely represent these characteristics while preserving all the useful structural information and how to efficiently and effectively map these structural characteristics to unknown, underlying properties and function(s). In recent decades, research in Artificial Intelligence, especially deep learning, has yielded impressive advancements in representing and generating high-dimensional data, such as sequence, image, and graph data, and in recent years, by representing molecules (and structures) as images or graphs, deep learning techniques have started to be utilized for molecular modeling; these techniques are drawing attention fast. Although encouraging results have been obtained in some applications, however, this line of research is in its infancy. The main aim of the Special Issue on “Deep Learning for Molecular Structure Modeling” is to be an open forum for researchers to share their findings via new techniques and applications. Specifically, we invite contributions that introduce new deep learning techniques towards better molecular (structure) characterization that takes into account spatial, network, and temporal properties under physical, biological, and chemical constraints. We also look forward to cutting-edge research on novel applications of deep learning for molecular modeling. Contributions to this issue, both in the form of original research or review articles, are welcome.

Prof. Dr. Amarda Shehu
Dr. Liang Zhao
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Molecules is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2000 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Molecular structure
  • Modeling and analysis
  • Deep learning
  • Representation learning
  • Deep generative models
  • Deep learning on graphs
  • Geometric deep learning
  • Artificial intelligence
  • Structural biology
  • Structure prediction and design

Published Papers (5 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Article
Learning the Fastest RNA Folding Path Based on Reinforcement Learning and Monte Carlo Tree Search
Molecules 2021, 26(15), 4420; https://doi.org/10.3390/molecules26154420 - 22 Jul 2021
Viewed by 302
Abstract
RNA molecules participate in many important biological processes, and they need to fold into well-defined secondary and tertiary structures to realize their functions. Like the well-known protein folding problem, there is also an RNA folding problem. The folding problem includes two aspects: structure [...] Read more.
RNA molecules participate in many important biological processes, and they need to fold into well-defined secondary and tertiary structures to realize their functions. Like the well-known protein folding problem, there is also an RNA folding problem. The folding problem includes two aspects: structure prediction and folding mechanism. Although the former has been widely studied, the latter is still not well understood. Here we present a deep reinforcement learning algorithms 2dRNA-Fold to study the fastest folding paths of RNA secondary structure. 2dRNA-Fold uses a neural network combined with Monte Carlo tree search to select residue pairing step by step according to a given RNA sequence until the final secondary structure is formed. We apply 2dRNA-Fold to several short RNA molecules and one longer RNA 1Y26 and find that their fastest folding paths show some interesting features. 2dRNA-Fold is further trained using a set of RNA molecules from the dataset bpRNA and is used to predict RNA secondary structure. Since in 2dRNA-Fold the scoring to determine next step is based on possible base pairings, the learned or predicted fastest folding path may not agree with the actual folding paths determined by free energy according to physical laws. Full article
(This article belongs to the Special Issue Deep Learning for Molecular Structure Modelling)
Show Figures

Figure 1

Article
Graph Neural Networks with Multiple Feature Extraction Paths for Chemical Property Estimation
Molecules 2021, 26(11), 3125; https://doi.org/10.3390/molecules26113125 - 24 May 2021
Viewed by 389
Abstract
Feature extraction is essential for chemical property estimation of molecules using machine learning. Recently, graph neural networks have attracted attention for feature extraction from molecules. However, existing methods focus only on specific structural information, such as node relationship. In this paper, we propose [...] Read more.
Feature extraction is essential for chemical property estimation of molecules using machine learning. Recently, graph neural networks have attracted attention for feature extraction from molecules. However, existing methods focus only on specific structural information, such as node relationship. In this paper, we propose a novel graph convolutional neural network that performs feature extraction with simultaneously considering multiple structures. Specifically, we propose feature extraction paths specialized in node, edge, and three-dimensional structures. Moreover, we propose an attention mechanism to aggregate the features extracted by the paths. The attention aggregation enables us to select useful features dynamically. The experimental results showed that the proposed method outperformed previous methods. Full article
(This article belongs to the Special Issue Deep Learning for Molecular Structure Modelling)
Show Figures

Figure 1

Article
Generative Adversarial Learning of Protein Tertiary Structures
Molecules 2021, 26(5), 1209; https://doi.org/10.3390/molecules26051209 - 24 Feb 2021
Cited by 1 | Viewed by 720
Abstract
Protein molecules are inherently dynamic and modulate their interactions with different molecular partners by accessing different tertiary structures under physiological conditions. Elucidating such structures remains challenging. Current momentum in deep learning and the powerful performance of generative adversarial networks (GANs) in complex domains, [...] Read more.
Protein molecules are inherently dynamic and modulate their interactions with different molecular partners by accessing different tertiary structures under physiological conditions. Elucidating such structures remains challenging. Current momentum in deep learning and the powerful performance of generative adversarial networks (GANs) in complex domains, such as computer vision, inspires us to investigate GANs on their ability to generate physically-realistic protein tertiary structures. The analysis presented here shows that several GAN models fail to capture complex, distal structural patterns present in protein tertiary structures. The study additionally reveals that mechanisms touted as effective in stabilizing the training of a GAN model are not all effective, and that performance based on loss alone may be orthogonal to performance based on the quality of generated datasets. A novel contribution in this study is the demonstration that Wasserstein GAN strikes a good balance and manages to capture both local and distal patterns, thus presenting a first step towards more powerful deep generative models for exploring a possibly very diverse set of structures supporting diverse activities of a protein molecule in the cell. Full article
(This article belongs to the Special Issue Deep Learning for Molecular Structure Modelling)
Show Figures

Figure 1

Article
Targeting the Initiator Protease of the Classical Pathway of Complement Using Fragment-Based Drug Discovery
Molecules 2020, 25(17), 4016; https://doi.org/10.3390/molecules25174016 - 03 Sep 2020
Viewed by 848
Abstract
The initiating protease of the complement classical pathway, C1r, represents an upstream and pathway-specific intervention point for complement-related autoimmune and inflammatory diseases. Yet, C1r-targeted therapeutic development is currently underrepresented relative to other complement targets. In this study, we developed a fragment-based drug discovery [...] Read more.
The initiating protease of the complement classical pathway, C1r, represents an upstream and pathway-specific intervention point for complement-related autoimmune and inflammatory diseases. Yet, C1r-targeted therapeutic development is currently underrepresented relative to other complement targets. In this study, we developed a fragment-based drug discovery approach using surface plasmon resonance (SPR) and molecular modeling to identify and characterize novel C1r-binding small-molecule fragments. SPR was used to screen a 2000-compound fragment library for binding to human C1r. This led to the identification of 24 compounds that bound C1r with equilibrium dissociation constants ranging between 160–1700 µM. Two fragments, termed CMP-1611 and CMP-1696, directly inhibited classical pathway-specific complement activation in a dose-dependent manner. CMP-1611 was selective for classical pathway inhibition, while CMP-1696 also blocked the lectin pathway but not the alternative pathway. Direct binding experiments mapped the CMP-1696 binding site to the serine protease domain of C1r and molecular docking and molecular dynamics studies, combined with C1r autoactivation assays, suggest that CMP-1696 binds within the C1r active site. The group of structurally distinct fragments identified here, along with the structure–activity relationship profiling of two lead fragments, form the basis for future development of novel high-affinity C1r-binding, classical pathway-specific, small-molecule complement inhibitors. Full article
(This article belongs to the Special Issue Deep Learning for Molecular Structure Modelling)
Show Figures

Figure 1

Article
Molecular Image-Based Prediction Models of Nuclear Receptor Agonists and Antagonists Using the DeepSnap-Deep Learning Approach with the Tox21 10K Library
Molecules 2020, 25(12), 2764; https://doi.org/10.3390/molecules25122764 - 15 Jun 2020
Cited by 4 | Viewed by 970
Abstract
The interaction of nuclear receptors (NRs) with chemical compounds can cause dysregulation of endocrine signaling pathways, leading to adverse health outcomes due to the disruption of natural hormones. Thus, identifying possible ligands of NRs is a crucial task for understanding the adverse outcome [...] Read more.
The interaction of nuclear receptors (NRs) with chemical compounds can cause dysregulation of endocrine signaling pathways, leading to adverse health outcomes due to the disruption of natural hormones. Thus, identifying possible ligands of NRs is a crucial task for understanding the adverse outcome pathway (AOP) for human toxicity as well as the development of novel drugs. However, the experimental assessment of novel ligands remains expensive and time-consuming. Therefore, an in silico approach with a wide range of applications instead of experimental examination is highly desirable. The recently developed novel molecular image-based deep learning (DL) method, DeepSnap-DL, can produce multiple snapshots from three-dimensional (3D) chemical structures and has achieved high performance in the prediction of chemicals for toxicological evaluation. In this study, we used DeepSnap-DL to construct prediction models of 35 agonist and antagonist allosteric modulators of NRs for chemicals derived from the Tox21 10K library. We demonstrate the high performance of DeepSnap-DL in constructing prediction models. These findings may aid in interpreting the key molecular events of toxicity and support the development of new fields of machine learning to identify environmental chemicals with the potential to interact with NR signaling pathways. Full article
(This article belongs to the Special Issue Deep Learning for Molecular Structure Modelling)
Show Figures

Figure 1

Back to TopTop