A Systematic Survey on Deep Generative Models for Graph Generation

Guo, Xiaojie; Zhao, Liang

doi:10.36227/techrxiv.12733037.v1

Cited by 12 publications

(9 citation statements)

References 76 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although most of these surveys have made a passing reference to some modern graph generators, this field requires individual attention due to its value and growing popularity. Recently, and somewhat concurrent to our work, Guo and Zhao [85] have exclusively reviewed this research field from totally different categorizations and perspectives than we do. Specifically, they have classified the existing methods based on properties of their generation process (e.g., whether the generation process is conditional or unconditional, one-shot or sequential, and etc.…”

Section: Introductionmentioning

confidence: 80%

“…), while we categorize them mainly according to their training objectives and model architectures in five classes (i.e., autoregressive, autoencoder-based, RL-based, adversarial, and flow-based) and identify key characteristics of the approaches in each category. Moreover, the authors of [85] neither provide supplementary resources related to DGGs (i.e., datasets and source codes) nor they investigate trends of techniques and applications. These are both included in this paper with additional analyses of each category's methods and their progression over time.…”

Section: Introductionmentioning

confidence: 99%

“…These are both included in this paper with additional analyses of each category's methods and their progression over time. We also review the most recent approaches, some of which are not covered in [85].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep Graph Generators: A Survey

et al. 2021

View full text Add to dashboard Cite

Deep generative models have achieved great success in areas such as image, speech, and natural language processing in the past few years. Thanks to the advances in graph-based deep learning, and in particular graph representation learning, deep graph generation methods have recently emerged with new applications ranging from discovering novel molecular structures to modeling social networks. This paper conducts a comprehensive survey on deep learning-based graph generation approaches and classifies them into five broad categories, namely, autoregressive, autoencoder-based, reinforcement learning-based, adversarial, and flow-based graph generators, providing the readers a detailed description of the methods in each class. We also present publicly available source codes, commonly used datasets, and the most widely utilized evaluation metrics. Finally, we review current trends and suggest future research directions based on the existing challenges.

show abstract

Section: Introductionmentioning

confidence: 80%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Graph Generators: A Survey

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In recent years, despite the growing attention towards graph-based generative models, the role of string-based generative models remains highly essential. Over the last five years, numerous surprisingly efficient graph-based generative models have been introduced, and de novo molecular design on graphs remains a hot topic in drug discovery [338,91]. The generation of molecular graphs is a promising direction, despite many unanswered questions and unaddressed issues.…”

Section: String-and Graph-based Modelsmentioning

confidence: 99%

“…Faez et al [72] conducted a survey on graph-based DL architectures for data generation. The review by Guo et al [91] discussed the challenges of graph-based generative models and categorized these models based on how molecular graphs are processed and produced. Zhu et al [338] summarized methods and applications of molecular graphs for de novo molecular design.…”

Section: Related Workmentioning

confidence: 99%

De novo Molecular Design using Deep Learning

Nguyen

View full text Add to dashboard Cite

Currently, the growth of data science, computer science, and artificial intelligence has transformed traditional drug discovery. The era of information has opened numerous opportunities for various research fields. The introduction of computer-aided stages (e.g., molecule generation, property prediction, virtual screening, etc.) into the drug discovery pipeline has immensely enhanced the success rate of finding promising molecules. Despite initial accomplishments, computer-aided drug discovery still needs to be significantly improved. Among the well-known topics in computer-aided drug discovery, 'de novo molecular design' is a highly focused topic that attracts a large number of researchers. De novo molecular design aims to excavate novel molecules from the huge chemical space that has not been fully exploited. Although there are various deep learning architectures proposed for molecule generation, each approach has certain limitations that need to be addressed. Additionally, since molecule generation is a random and non-directional process, finding drug candidates with desired properties from billions of molecules is almost infeasible. To tackle this problem, several optimization techniques were utilized to direct the generative model to produce `molecule of interest'. However, the property-optimized process restricts the 'creativity' of the generative model. Furthermore, it is a fact that not every desired property can be optimized because of insufficient data, and optimization-driving generation is computationally expensive. In such cases, using Quantitative Structure-Activity Relationship (QSAR) models is an alternative solution for identifying molecules with desired properties.The overall goal of this thesis is to develop a generative model and a series of QSAR models for drug discovery. The generative model is used to produce novel molecules, while the QSAR models are used to virtually filter the molecules with desired properties. To achieve this goal, a range of computational techniques and interdisciplinary knowledge are employed in this thesis. First, we conducted a critical review of existing molecular representations, generative models, and property prediction models. The review is highly essential to providing readers with a fundamental understanding of de novo molecular design. The review analyzes the pros and cons of each molecular representation and summarizes the present development and challenges of molecular generation and property prediction tasks. Second, we investigated a novel deep learning architecture for de novo molecular design. The architecture is designed to process graph-structure data. The generative model developed using the proposed architecture can produce hypothetical molecules with high novelty and diversity. Experimental results indicated that our generative model can create drug-like molecules varying in size, scaffold, and properties.Third, we proposed two novel deep learning architectures for molecular property prediction. These two architectures, including the Residual Graph Attention (ResGAT) Network and the Graph Convolution-Attention Network (GCoAtNet), are designed to process graph-structure data. Our findings demonstrated that ResGAT achieved competitive performance while GCoAtNet achieved higher performance compared to state-of-the-art architectures. Our models were benchmarked against these state-of-the-art models on nine molecular datasets. Finally, we used these proposed architectures to construct a generative model and two QSAR models. The generative model was driven to produce a large number of hypothetical molecules. Subsequently, these molecules were virtually screened to eliminate those with drug-induced liver injury (property 1}) and Cytochrome-P450-inhibitory (property 2) activities. For each property, we developed two QSAR models that can independently identify molecules with desired properties. The intersection set of molecules suggested by these two models was considered a short list of potential drug candidates. These shortlisted molecules can be sent to the chemistry lab for further investigation, i.e., structural optimization and modification, synthesis, and evaluation. The results demonstrated that these computer-designed molecules are synthesizable and suitable for further research.

show abstract