A convolutional neural network and graph convolutional network-based method for predicting the classification of anatomical therapeutic chemicals

PDATC-NCPMKL: Predicting drug's Anatomical Therapeutic Chemical (ATC) codes based on network consistency projection and multiple kernel learning

The development and discovery of new drugs is time-consuming and needs lots of human and material resources. Therefore, discovery of novel effects of existing drugs is an important alternative way, which can accelerate the process of designing "new" drugs. The anatomical Therapeutic Chemical (ATC) classification system recommended by World Health Organization (WHO) is a basic research area in this regard. A novel ATC code of an existing drug suggests its novel effects. Some computational models have been proposed, which can predict the drug-ATC code associations. However, their performance is not very high. There still exist spaces for improvement. In this study, a new recommendation system (named PDATC-NCPMKL), which incorporated network consistency projection and multi-kernel learning, was designed to identify drug-ATC code associations. For drugs or ATC codes, several kernels were constructed, which were fused by a multiple kernel learning method and an additional kernel integration scheme. To enhance the performance, the drug-ATC code association adjacency matrix was reformulated by a variant of weighted K nearest known neighbors (WKNKN). The reformulated adjacency matrix, drug and ATC code kernels were fed into network consistency projection to generate the association score matrix. The proposed recommendation system was tested on the ATC codes at the second, third and fourth levels in drug ATC classification system using ten-fold cross-validation. The results indicated that all AUROC and AUPR values were close to or exceeded 0.96. Such performance was higher than some existing computational models. Some additional tests were conducted to prove the utility of adjacency matrix reformulation and to analyze the importance of drug and ATC code kernels.

A GraphSAGE-based model with fingerprints only to predict drug-drug interactions

Drugs are an effective way to treat various diseases. Some diseases are so complicated that the effect of a single drug for such diseases is limited, which has led to the emergence of combination drug therapy. The use multiple drugs to treat these diseases can improve the drug efficacy, but it can also bring adverse effects. Thus, it is essential to determine drug-drug interactions (DDIs). Recently, deep learning algorithms have become popular to design DDI prediction models. However, most deep learning-based models need several types of drug properties, inducing the application problems for drugs without these properties. In this study, a new deep learning-based model was designed to predict DDIs. For wide applications, drugs were first represented by commonly used properties, referred to as fingerprint features. Then, these features were perfectly fused with the drug interaction network by a type of graph convolutional network method, GraphSAGE, yielding high-level drug features. The inner product was adopted to score the strength of drug pairs. The model was evaluated by 10-fold cross-validation, resulting in an AUROC of 0.9704 and AUPR of 0.9727. Such performance was better than the previous model which directly used drug fingerprint features and was competitive compared with some other previous models that used more drug properties. Furthermore, the ablation tests indicated the importance of the main parts of the model, and we analyzed the strengths and limitations of a model for drugs with different degrees in the network. This model identified some novel DDIs that may bring expected benefits, such as the combination of PEA and cannabinol that may produce better effects. DDIs that may cause unexpected side effects have also been discovered, such as the combined use of WIN 55,212-2 and cannabinol. These DDIs can provide novel insights for treating complex diseases or avoiding adverse drug events.

Improved multi-label classifiers for predicting protein subcellular localization

Protein functions are closely related to their subcellular locations. At present, the prediction of protein subcellular locations is one of the most important problems in protein science. The evident defects of traditional methods make it urgent to design methods with high efficiency and low costs. To date, lots of computational methods have been proposed. However, this problem is far from being completely solved. Recently, some multi-label classifiers have been proposed to identify subcellular locations of human, animal, Gram-negative bacterial and eukaryotic proteins. These classifiers adopted the protein features derived from gene ontology information. Although they provided good performance, they can be further improved by adopting more powerful machine learning algorithms. In this study, four improved multi-label classifiers were set up for identification of subcellular locations of the above four protein types. The random k-labelsets (RAKEL) algorithm was used to tackle proteins with multiple locations, and random forest was used as the basic prediction engine. All classifiers were tested by jackknife test, indicating their high performance. Comparisons with previous classifiers further confirmed the superiority of the proposed classifiers.

LDS-CNN: a deep learning framework for drug-target interactions prediction based on large-scale drug screening

Background

Drug-target interaction (DTI) is a vital drug design strategy that plays a significant role in many processes of complex diseases and cellular events. In the face of challenges such as extensive protein data and experimental costs, it is suggested to apply bioinformatics approaches to exploit potential interactions to design new targeted medications. Different data and interaction types bring difficulties to study involving incompatible and heterology formats. The analysis of drug-target interactions in a comprehensive and unified model is a significant challenge.

Method

Here, we propose a general method for predicting interactions between small-molecule drugs and protein targets, Large-scale Drug target Screening Convolutional Neural Network (LDS-CNN), which used unified encoding to achieve the calculation of the different data formats in an integrated model to realize feature abstraction and potential object prediction.

Result

On 898,412 interaction data involving 1683 small-molecule compounds and 14,350 human proteins from 8.8 billion records, the proposed method achieved an area under the curve (AUC) of 0.96, an area under the precision-recall curve (AUPRC) of 0.95, and an accuracy of 90.13%. The experimental results illustrated that the proposed method attained high accuracy on the test set, indicating its high predictive ability in drug-target interaction prediction. LDS-CNN is effective for the prediction of large-scale datasets and datasets composed of data with different formats.

Conclusion

In this study, we propose a DTI prediction method to solve the problems of unified encoding of large-scale data in multiple formats. It provides a feasible way to efficiently abstract the features among different types of drug-related data, thus reducing experimental costs and time consumption. The proposed method can be used to identify potential drug targets and candidates for the treatment of complex diseases. This work provides a reference for DTI to process large-scale data and different formats with deep learning methods and provides certain suggestions for future research.

Cross-view contrastive representation learning approach to predicting DTIs via integrating multi-source information

Drug-target interaction (DTI) prediction serves as the foundation of new drug findings and drug repositioning. For drugs/targets, the sequence data contains the biological structural information, while the heterogeneous network contains the biochemical functional information. These two types of information describe different aspects of drugs and targets. Due to the complexity of DTI machinery, it is necessary to learn the representation from multiple perspectives. We hereby try to design a way to leverage information from multi-source data to the maximum extent and find a strategy to fuse them. To address the above challenges, we propose a model, named MOVE (short for integrating multi-source information for predicting DTI via cross-view contrastive learning), for learning comprehensive representations of each drug and target from multi-source data. MOVE extracts information from the sequence view and the network view, then utilizes a fusion module with auxiliary contrastive learning to facilitate the fusion of representations. Experimental results on the benchmark dataset demonstrate that MOVE is effective in DTI prediction.

RNPredATC: A Deep Residual Learning-Based Model With Applications to the Prediction of Drug-ATC Code Association

The Anatomical Therapeutic Chemical (ATC) classification system, designated by the World Health Organization Collaborating Center (WHOCC), has been widely used in drug screening, repositioning, and similarity research. The ATC classification system assigns different codes to drugs according to the organ or system on which they act and/or their therapeutic and chemical characteristics. Correctly identifying the potential ATC codes for drugs can accelerate drug development and reduce the cost of experiments. Several classifiers have been proposed in this regard. However, they lack of ability to learn basic features from sparsely known drug-ATC code associations. Therefore, there is an urgent need for novel computational methods to precisely predict potential drug-ATC code associations in multiple levels of the ATC classification system based on known associations between drugs and ATC codes. In this paper, we provide a novel end-to-end model, so-called RNPredATC, to predict potential drug-ATC code associations in five ATC classification levels. RNPredATC can extract dense feature vectors from sparsely known drug-ATC code associations and reduce the impact from the degradation problem by a novel deep residual learning. We extensively compare our method with some state-of-the-art methods, including NetPredATC, SPACE, and some multi-label-based methods. Our experimental results show that RNPredATC achieves better performances in five-fold and ten-fold cross validations. Furthermore, the visualization analysis of hidden layers and case studies of predicted associations at the fifth ATC classification level confirm that RNPredATC can effectively identify the potential ATC codes of drugs.

Identifying the serious clinical outcomes of adverse reactions to drugs by a multi-task deep learning framework

Adverse Drug Reactions (ADRs) have a direct impact on human health. As continuous pharmacovigilance and drug monitoring prove to be costly and time-consuming, computational methods have emerged as promising alternatives. However, most existing computational methods primarily focus on predicting whether or not the drug is associated with an adverse reaction and do not consider the core issue of drug benefit-risk assessment-whether the treatment outcome is serious when adverse drug reactions occur. To this end, we categorize serious clinical outcomes caused by adverse reactions to drugs into seven distinct classes and present a deep learning framework, so-called GCAP, for predicting the seriousness of clinical outcomes of adverse reactions to drugs. GCAP has two tasks: one is to predict whether adverse reactions to drugs cause serious clinical outcomes, and the other is to infer the corresponding classes of serious clinical outcomes. Experimental results demonstrate that our method is a powerful and robust framework with high extendibility. GCAP can serve as a useful tool to successfully address the challenge of predicting the seriousness of clinical outcomes stemming from adverse reactions to drugs.

A Self-attention Graph Convolutional Network for Precision Multi-tumor Early Diagnostics with DNA Methylation Data

DNA methylation-based precision tumor early diagnostics is emerging as state-of-the-art technology that could capture early cancer signs 3 ~ 5 years in advance, even for clinically homogenous groups. Presently, the sensitivity of early detection for many tumors is ~ 30%, which needs significant improvement. Nevertheless, based on the genome-wide DNA methylation data, one could comprehensively characterize tumors' entire molecular genetic landscape and their subtle differences. Therefore, novel high-performance methods must be modeled by considering unbiased information using excessively available DNA methylation data. To fill this gap, we have designed a computational model involving a self-attention graph convolutional network and multi-class classification support vector machine to identify the 11 most common cancers using DNA methylation data. The self-attention graph convolutional network automatically learns key methylation sites in a data-driven way. Then, multi-tumor early diagnostics is realized by training a multi-class classification support vector machine based on the selected methylation sites. We evaluated our model's performance through several data sets of experiments, and our results demonstrate the effectiveness of the selected key methylation sites, which are highly relevant for blood diagnosis. The pipeline of the self-attention graph convolutional network based computational framework.

AttentionDTA: Drug-Target Binding Affinity Prediction by Sequence-Based Deep Learning With Attention Mechanism

The identification of drug-target relations (DTRs) is substantial in drug development. A large number of methods treat DTRs as drug-target interactions (DTIs), a binary classification problem. The main drawback of these methods are the lack of reliable negative samples and the absence of many important aspects of DTR, including their dose dependence and quantitative affinities. With increasing number of publications of drug-protein binding affinity data recently, DTRs prediction can be viewed as a regression problem of drug-target affinities (DTAs) which reflects how tightly the drug binds to the target and can present more detailed and specific information than DTIs. The growth of affinity data enables the use of deep learning architectures, which have been shown to be among the state-of-the-art methods in binding affinity prediction. Although relatively effective, due to the black-box nature of deep learning, these models are less biologically interpretable. In this study, we proposed a deep learning-based model, named AttentionDTA, which uses attention mechanism to predict DTAs. Different from the models using 3D structures of drug-target complexes or graph representation of drugs and proteins, the novelty of our work is to use attention mechanism to focus on key subsequences which are important in drug and protein sequences when predicting its affinity. We use two separate one-dimensional Convolution Neural Networks (1D-CNNs) to extract the semantic information of drug's SMILES string and protein's amino acid sequence. Furthermore, a two-side multi-head attention mechanism is developed and embedded to our model to explore the relationship between drug features and protein features. We evaluate our model on three established DTA benchmark datasets, Davis, Metz, and KIBA. AttentionDTA outperforms the state-of-the-art deep learning methods under different evaluation metrics. The results show that the attention-based model can effectively extract protein features related to drug information and drug features related to protein information to better predict drug target affinities. It is worth mentioning that we test our model on IC50 dataset, which provides the binding sites between drugs and proteins, to evaluate the ability of our model to locate binding sites. Finally, we visualize the attention weight to demonstrate the biological significance of the model. The source code of AttentionDTA can be downloaded from https://github.com/zhaoqichang/AttentionDTA_TCBB.

A model with deep analysis on a large drug network for drug classification

Drugs are an important means to treat various diseases. They are classified into several classes to indicate their properties and effects. Those in the same class always share some important features. The Kyoto Encyclopedia of Genes and Genomes (KEGG) DRUG recently reported a new drug classification system that classifies drugs into 14 classes. Correct identification of the class for any possible drug-like compound is helpful to roughly determine its effects for a particular type of disease. Experiments could be conducted to confirm such latent effects, thus accelerating the procedures for discovering novel drugs. In this study, this classification system was investigated. A classification model was proposed to assign one of the classes in the system to any given drug for the first time. Different from traditional fingerprint features, which indicated essential drug properties alone and were very popular in investigating drug-related problems, drugs were represented by novel features derived from a large drug network via a well-known network embedding algorithm called Node2vec. These features abstracted the drug associations generated from their essential properties, and they could overview each drug with all drugs as background. As class sizes were of great differences, synthetic minority over-sampling technique (SMOTE) was employed to tackle the imbalance problem. A balanced dataset was fed into the support vector machine to build the model. The 10-fold cross-validation results suggested the excellent performance of the model. This model was also superior to models using other drug features, including those generated by another network embedding algorithm and fingerprint features. Furthermore, this model provided more balanced performance across all classes than that without SMOTE.

Identifying the kind behind SMILES-anatomical therapeutic chemical classification using structure-only representations

Anatomical Therapeutic Chemical (ATC) classification for compounds/drugs plays an important role in drug development and basic research. However, previous methods depend on interactions extracted from STITCH dataset which may make it depend on lab experiments. We present a pilot study to explore the possibility of conducting the ATC prediction solely based on the molecular structures. The motivation is to eliminate the reliance on the costly lab experiments so that the characteristics of a drug can be pre-assessed for better decision-making and effort-saving before the actual development. To this end, we construct a new benchmark consisting of 4545 compounds which is with larger scale than the one used in previous study. A light-weight prediction model is proposed. The model is with better explainability in the sense that it is consists of a straightforward tokenization that extracts and embeds statistically and physicochemically meaningful tokens, and a deep network backed by a set of pyramid kernels to capture multi-resolution chemical structural characteristics. Its efficacy has been validated in the experiments where it outperforms the state-of-the-art methods by 15.53% in accuracy and by 69.66% in terms of efficiency. We make the benchmark dataset, source code and web server open to ease the reproduction of this study.

DACPGTN: Drug ATC Code Prediction Method Based on Graph Transformer Network for Drug Discovery

The Anatomical Therapeutic Chemical (ATC) classification system is a drug classification scheme proposed by the World Health Organization, which is widely used for drug screening, repositioning, and similarity research. The ATC system assigns different ATC codes to drugs based on their anatomy, pharmacological, therapeutics and chemical properties. Predicting the ATC code of a given drug helps to understand the indication and potential toxicity of the drug, thus promoting its use in the therapeutic phase and accelerating its development. In this article, we propose an end-to-end model DACPGTN to predict the ATC code for the given drug. DACPGTN constructs composite features of drugs, diseases and targets by applying diverse biomedical information. Inspired by the application of Graph Transformer Network, we learn potential novel interactions among drugs diseases and targets from the known interactions to construct drug-target-disease heterogeneous networks containing comprehensive interaction information. Based on the constructed composite features and learned heterogeneous networks, we employ graph convolution network to generate the embedding of drug nodes, which are further used for the multi-label learning tasks in drug discovery. Experiments on the benchmark datasets demonstrate that the proposed DACPGTN model can achieve better prediction performance than the existing methods. The source codes of our method are available at https://github.com/Szhgege/DACPGTN.

Neural networks for anatomical therapeutic chemical (ATC) classification

Purpose

Automatic anatomical therapeutic chemical (ATC) classification is progressing at a rapid pace because of its potential in drug development. Predicting an unknown compound's therapeutic and chemical characteristics in terms of how it affects multiple organs and physiological systems makes automatic ATC classification a vital yet challenging multilabel problem. The aim of this paper is to experimentally derive an ensemble of different feature descriptors and classifiers for ATC classification that outperforms the state-of-the-art.

Design/methodology/approach

The proposed method is an ensemble generated by the fusion of neural networks (i.e. a tabular model and long short-term memory networks (LSTM)) and multilabel classifiers based on multiple linear regression (hMuLab). All classifiers are trained on three sets of descriptors. Features extracted from the trained LSTMs are also fed into hMuLab. Evaluations of ensembles are compared on a benchmark data set of 3883 ATC-coded pharmaceuticals taken from KEGG, a publicly available drug databank.

Findings

Experiments demonstrate the power of the authors’ best ensemble, EnsATC, which is shown to outperform the best methods reported in the literature, including the state-of-the-art developed by the fast.ai research group. The MATLAB source code of the authors’ system is freely available to the public at https://github.com/LorisNanni/Neural-networks-for-anatomical-therapeutic-chemical-ATC-classification.

Originality/value

This study demonstrates the power of extracting LSTM features and combining them with ATC descriptors in ensembles for ATC classification.