A new model for learning in graph domains

Gori, Marco; Monfardini, Gabriele; Scarselli, Franco

doi:10.1109/ijcnn.2005.1555942

Cited by 1,393 publications

(936 citation statements)

References 7 publications

Supporting

Mentioning

849

Contrasting

Unclassified

Order By: Relevance

“…More precisely, they encoded the distribution of each individual features by a n-bits histogram. From another perspective, several authors [47,48,49] proposed to learn a graph-level embeddings by creating a virtual "supernode", that is connected to all other nodes via a particular type of edge. This virtual node is updated like real nodes, and the final representation of this virtual node is used as the graph representation.…”

Section: Molecular Graph Encodermentioning

confidence: 99%

Evaluation of network architecture and data augmentation methods for deep learning in chemogenomics

Playe

Stoven

2019

Preprint

View full text Add to dashboard Cite

Among virtual screening methods that have been developed to facilitate the drug discovery process, chemogenomics presents the particularity to tackle the question of predicting ligands for proteins, at at scales both in the protein and chemical spaces. Therefore, in addition to to predict drug candidates for a given therapeutic protein target, like more classical ligand-based or receptor-based methods do, chemogenomics can also predict off-targets at the proteome level, and therefore, identify potential side-effects or drug repositioning opportunities. In this study, we study and compare machinelearning and deep learning approaches for chemogenomics, that are applicable to screen large sets of compounds against large sets of druggable proteins. State-of-the-art drug chemogenomics methods rely on expert-based chemical and protein descriptors or similarity measures. The recent development of deep learning approaches enabled to design algorithms that learn numerical abstract representations of molecular graphs and protein sequences in an end-to-end fashion, i.e., so that the learnt features optimise the objective function of the drug-target interaction prediction task. In this paper, we address drug-target interaction prediction at the druggable proteome-level, with what we define as the chemogenomic neuron network. This network consists of a feed-forward neuron network taking as input the combination of molecular and protein representations learnt by molecular graph and protein sequence encoders. We first propose a standard formulation of this chemogenomic neuron network. Then, we compare the performances of the standard chemogenomic network to reference deep learning or shallow (machine-learning without deep learning) methods. In particular, we show that such a representation learning approach is competitive with state-of-the-art chemogenomics with shallow methods, but not ultimately superior. We evaluate the most promising neuron network architectures and data augmentation techniques, such as multi-view and transfer learning, to improve the prediction performance of the chemogenomic network. Our results shed new insights on the design of chemogenomics approaches based on representation learning algorithms. Most importantly, we conclude from our observations that a promising research direction is to integrate heterogeneous sources of data such as various bioactivity datasets, or independently, multiple molecule and protein attribute views, instead of focusing on sophisticated, yet intuitively relevant, encoder's neuron network architecture.

show abstract

Section: Molecular Graph Encodermentioning

confidence: 99%

Evaluation of network architecture and data augmentation methods for deep learning in chemogenomics

Playe

Stoven

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…It is worth to mention that even if the gradient could be computed also by applying the standard backpropagation through time algorithm (Werbos 1990) to the encoding network, however the procedure adopted by GNNs is faster and uses less memory by exploiting Almeida-Pineda algorithm peculiarities. More details can be found in Gori et al (2005), Scarselli et al (2009b).…”

Section: Graph Neural Networkmentioning

confidence: 99%

“…They extend support vector machines (Kondor and Lafferty 2002;Gärtner 2003), neural networks (Sperduti and Starita 1997;Frasconi et al 1998;Gori et al 2005) and SOMs to structured data. The main idea underlying those methods is to automatically obtain an internal flat representation of the symbolic and subsymbolic information collected in the graphs.…”

Section: Introductionmentioning

confidence: 99%

Neural networks for relational learning: an experimental comparison

et al. 2010

View full text Add to dashboard Cite

In the last decade, connectionist models have been proposed that can process structured information directly. These methods, which are based on the use of graphs for the representation of the data and the relationships within the data, are particularly suitable for handling relational learning tasks. In this paper, two recently proposed architectures of this kind, i.e. Graph Neural Networks (GNNs) and Relational Neural Networks (RelNNs), are compared and discussed, along with their corresponding learning schemes. The goal is to evaluate the performance of these methods on benchmarks that are commonly used by the relational learning community. Moreover, we also aim at reporting differences in the behavior of the two models, in order to gain insights on possible extensions of the approaches. Since RelNNs have been developed with the specific task of learning aggregate functions in mind, some experiments are run considering that particular task. In addition, we carry out more general experiments on the mutagenesis and the biodegradability datasets, on which several other relational learners have been evaluated. The experimental results are promising and suggest that RelNNs and GNNs can be a viable approach for learning on relational data.

show abstract

“…Graph Neural Networks (GNNs) have been recently proposed [17], [18] as an evolution of Recursive Neural Networks and have been conceived to fully implement the framework presented in Sect. III, in particular Eqs.…”

Section: B Graph Neural Networkmentioning

confidence: 99%

A Comparison between Recursive Neural Networks and Graph Neural Networks

Massa

Monfardini

Sarti

et al. 2006

The 2006 IEEE International Joint Conference on Neural Network Proceedings

Self Cite

View full text Add to dashboard Cite

Recursive neural networks (RNNs) and graph neural networks (GNNs) are two connectionist models that can directly process graphs. RNNs and GNNs exploit a similar processing framework, but they can be applied to different input domains. RNNs require the input graphs to be directed and acyclic, whereas GNNs can process any kind of graphs. The aim of this paper consists in understanding whether such a difference affects the behaviour of the models on a real application. An experimental comparison on an image classification problem is presented, showing that GNNs outperforms RNNs. Moreover the main differences between the models are also discussed w.r.t. their input domains, their approximation capabilities and their learning algorithms

show abstract

A new model for learning in graph domains

Cited by 1,393 publications

References 7 publications

Evaluation of network architecture and data augmentation methods for deep learning in chemogenomics

Evaluation of network architecture and data augmentation methods for deep learning in chemogenomics

Neural networks for relational learning: an experimental comparison

A Comparison between Recursive Neural Networks and Graph Neural Networks

Contact Info

Product

Resources

About