ATC-NLSP: Prediction of the Classes of Anatomical Therapeutic Chemicals Using a Network-Based Label Space Partition Method

Multimodal Machine Learning-Based Ductal Carcinoma in situ Prediction from Breast Fibromatosis

Objective

To develop a clinical-radiomics model using a multimodal machine learning method for distinguishing ductal carcinoma in situ (DCIS) from breast fibromatosis.

Methods

The clinical factors, ultrasound features, and related ultrasound images of 306 patients (198 DCIS patients) were retrospectively collected. Patients in the development and validation cohort were 184 and 122, respectively. The independent clinical and ultrasound factors identified by the multivariable logistic regression analysis were used for the clinical-ultrasound model construction. Then, the region of interest of breast lesions was delineated and radiomics features were extracted. Six machine learning algorithms were trained to develop a radiomics model. The algorithm with higher and more stable prediction ability was chosen to convert the output of the results into the Radscore. Further, the independent clinical predictors and Radscore were enrolled into the logistic regression analysis to generate a combined clinical-radiomics model. The receiver operating characteristic curve analysis, DeLong test, and decision curve analysis were adopted to compare the prediction ability and clinical efficacy of three different models.

Results

Among the six classifiers, logistic regression model was selected as the final radiomics model. Besides, the combined clinical-radiomics model exhibited a superior ability in distinguishing DCIS from breast fibromatosis to the clinical-ultrasound model and the radiomics model.

Conclusion

The combined model by integrating clinical-ultrasound factors and radiomics features performed well in predicting DCIS, which might promote prompt interventions to improve the early diagnosis and prognosis of the patients.

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.

Multitask joint learning with graph autoencoders for predicting potential MiRNA-drug associations

The occurrence of many diseases is associated with miRNA abnormalities. Predicting potential drug-miRNA associations is of great importance for both disease treatment and new drug discovery. Most computation-based approaches learn one task at a time, ignoring the information contained in other tasks in the same domain. Multitask learning can effectively enhance the prediction performance of a single task by extending the valid information of related tasks. In this paper, we presented a multitask joint learning framework (MTJL) with a graph autoencoder for predicting the associations between drugs and miRNAs. First, we combined multiple pieces of information to construct a high-quality similarity network of both drugs and miRNAs and then used a graph autoencoder (GAE) to learn their embedding representations separately. Second, to further improve the embedding quality of drugs, we added an auxiliary task to classify drugs using the learned representations. Finally, the embedding representations of drugs and miRNAs were linearly transformed to obtain the predictive association scores between them. A comparison with other state-of-the-art models shows that MTJL has the best prediction performance, and ablation experiments show that the auxiliary task can enhance the embedding quality and improve the robustness of the model. In addition, we show that MTJL has high utility in predicting potential associations between drugs and miRNAs by conducting two case studies.

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.

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.

Small Molecular Drug Screening Based on Clinical Therapeutic Effect

Virtual screening can significantly save experimental time and costs for early drug discovery. Drug multi-classification can speed up virtual screening and quickly predict the most likely class for a drug. In this study, 1019 drug molecules with actual therapeutic effects are collected from multiple databases and documents, and molecular sets are grouped according to therapeutic effect and mechanism of action. Molecular descriptors and molecular fingerprints are obtained through SMILES to quantify molecular structures. After using the Kennard–Stone method to divide the data set, a better combination can be obtained by comparing the combined results of five classification algorithms and a fusion method. Furthermore, for a specific data set, the model with the best performance is used to predict the validation data set. The test set shows that prediction accuracy can reach 0.862 and kappa coefficient can reach 0.808. The highest classification accuracy of the validation set is 0.873. The more reliable molecular set has been found, which could be used to predict potential attributes of unknown drug compounds and even to discover new use for old drugs. We hope this research can provide a reference for virtual screening of multiple classes of drugs at the same time in the future.

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.

SuperPred 3.0: drug classification and target prediction-a machine learning approach

Since the last published update in 2014, the SuperPred webserver has been continuously developed to offer state-of-the-art models for drug classification according to ATC classes and target prediction. For the first time, a thoroughly filtered ATC dataset, that is suitable for accurate predictions, is provided along with detailed information on the achieved predictions. This aims to overcome the challenges in comparing different published prediction methods, since performance can vary greatly depending on the training dataset used. Additionally, both ATC and target prediction have been reworked and are now based on machine learning models instead of overall structural similarity, stressing the importance of functional groups for the mechanism of action of small molecule substances. Additionally, the dataset for the target prediction has been extensively filtered and is no longer only based on confirmed binders but also includes non-binding substances to reduce false positives. Using these methods, accuracy for the ATC prediction could be increased by almost 5% to 80.5% compared to the previous version, and additionally the scoring function now offers values which are easily assessable at first glance. SuperPred 3.0 is publicly available without the need for registration at: https://prediction.charite.de/index.php.

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.

Identification of protein functions in mouse with a label space partition method

Protein is very important for almost all living creatures because it participates in most complicated and essential biological processes. Determining the functions of given proteins is one of the most essential problems in protein science. Such determination can be conducted through traditional experiments. However, the experimental methods are always time-consuming and of high costs. In recent years, computational methods give useful aids for identification of protein functions. This study presented a new multi-label classifier for identifying functions of mouse proteins. Due to the number of functional types, which were termed as labels in the classification procedure, a label space partition method was employed to divide labels into some partitions. On each partition, a multi-label classifier was constructed. The classifiers based on all partitions were integrated in the proposed classifier. The cross-validation results proved that the proposed classifier was of good performance. Classifiers with label partition were superior to those without label partition or with random label partition.

MDF-SA-DDI: predicting drug-drug interaction events based on multi-source drug fusion, multi-source feature fusion and transformer self-attention mechanism

One of the main problems with the joint use of multiple drugs is that it may cause adverse drug interactions and side effects that damage the body. Therefore, it is important to predict potential drug interactions. However, most of the available prediction methods can only predict whether two drugs interact or not, whereas few methods can predict interaction events between two drugs. Accurately predicting interaction events of two drugs is more useful for researchers to study the mechanism of the interaction of two drugs. In the present study, we propose a novel method, MDF-SA-DDI, which predicts drug-drug interaction (DDI) events based on multi-source drug fusion, multi-source feature fusion and transformer self-attention mechanism. MDF-SA-DDI is mainly composed of two parts: multi-source drug fusion and multi-source feature fusion. First, we combine two drugs in four different ways and input the combined drug feature representation into four different drug fusion networks (Siamese network, convolutional neural network and two auto-encoders) to obtain the latent feature vectors of the drug pairs, in which the two auto-encoders have the same structure, and their main difference is the number of neurons in the input layer of the two auto-encoders. Then, we use transformer blocks that include self-attention mechanism to perform latent feature fusion. We conducted experiments on three different tasks with two datasets. On the small dataset, the area under the precision-recall-curve (AUPR) and F1 scores of our method on task 1 reached 0.9737 and 0.8878, respectively, which were better than the state-of-the-art method. On the large dataset, the AUPR and F1 scores of our method on task 1 reached 0.9773 and 0.9117, respectively. In task 2 and task 3 of two datasets, our method also achieved the same or better performance as the state-of-the-art method. More importantly, the case studies on five DDI events are conducted and achieved satisfactory performance. The source codes and data are available at https://github.com/ShenggengLin/MDF-SA-DDI.

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

MOTIVATION

The Anatomical Therapeutic Chemical (ATC) system is an official classification system established by the World Health Organization for medicines. Correctly assigning ATC classes to given compounds is an important research problem in drug discovery, which can not only discover the possible active ingredients of the compounds, but also infer theirs therapeutic, pharmacological and chemical properties.

RESULTS

In this article, we develop an end-to-end multi-label classifier called CGATCPred to predict 14 main ATC classes for given compounds. In order to extract rich features of each compound, we use the deep Convolutional Neural Network and shortcut connections to represent and learn the seven association scores between the given compound and others. Moreover, we construct the correlation graph of ATC classes and then apply graph convolutional network on the graph for label embedding abstraction. We use all label embedding to guide the learning process of compound representation. As a result, by using the Jackknife test, CGATCPred obtain reliable Aiming of 81.94%, Coverage of 82.88%, Accuracy 80.81%, Absolute True 76.58% and Absolute False 2.75%, yielding significantly improvements compared to exiting multi-label classifiers.

AVAILABILITY AND IMPLEMENTATION

The codes of CGATCPred are available at https://github.com/zhc940702/CGATCPred and https://zenodo.org/record/4552917.

Deep fusion learning facilitates anatomical therapeutic chemical recognition in drug repurposing and discovery

The advent of large-scale biomedical data and computational algorithms provides new opportunities for drug repurposing and discovery. It is of great interest to find an appropriate data representation and modeling method to facilitate these studies. The anatomical therapeutic chemical (ATC) classification system, proposed by the World Health Organization (WHO), is an essential source of information for drug repurposing and discovery. Besides, computational methods are applied to predict drug ATC classification. We conducted a systematic review of ATC computational prediction studies and revealed the differences in data sets, data representation, algorithm approaches, and evaluation metrics. We then proposed a deep fusion learning (DFL) framework to optimize the ATC prediction model, namely DeepATC. The methods based on graph convolutional network, inferring biological network and multimodel attentive fusion network were applied in DeepATC to extract the molecular topological information and low-dimensional representation from the molecular graph and heterogeneous biological networks. The results indicated that DeepATC achieved superior model performance with area under the curve (AUC) value at 0.968. Furthermore, the DFL framework was performed for the transcriptome data-based ATC prediction, as well as another independent task that is significantly relevant to drug discovery, namely drug-target interaction. The DFL-based model achieved excellent performance in the above-extended validation task, suggesting that the idea of aggregating the heterogeneous biological network and node's (molecule or protein) self-topological features will bring inspiration for broader drug repurposing and discovery research.

iATC-FRAKEL: a simple multi-label web server for recognizing anatomical therapeutic chemical classes of drugs with their fingerprints only

MOTIVATION

Anatomical therapeutic chemical (ATC) classification system is very important for drug utilization and studies. Correct prediction of the 14 classes in the first level for given drugs is an essential problem for the study on such system. Several multi-label classifiers have been proposed in this regard. However, only two of them provided the web servers and their performance was not very high. On the other hand, although some rest classifiers can provide better performance, they were built based on some prior knowledge on drugs, such as information of chemical-chemical interaction and chemical ontology, leading to limited applications. Furthermore, provided codes of these classifiers are almost inaccessible for pharmacologists.

RESULTS

In this study, we built a simple web server, namely iATC-FRAKEL. This web server only required the SMILES format of drugs as input and extracted their fingerprints for making prediction. The performance of the iATC-FRAKEL was much higher than all existing web servers and was comparable to the best multi-label classifier but had much wider applications. Such web server can be visited at http://cie.shmtu.edu.cn/iatc/index.

AVAILABILITY AND IMPLEMENTATION

The web server is available at http://cie.shmtu.edu.cn/iatc/index.

CONTACT

chen_lei1@163.com.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

T4SE-XGB: Interpretable Sequence-Based Prediction of Type IV Secreted Effectors Using eXtreme Gradient Boosting Algorithm

Type IV secreted effectors (T4SEs) can be translocated into the cytosol of host cells via type IV secretion system (T4SS) and cause diseases. However, experimental approaches to identify T4SEs are time- and resource-consuming, and the existing computational tools based on machine learning techniques have some obvious limitations such as the lack of interpretability in the prediction models. In this study, we proposed a new model, T4SE-XGB, which uses the eXtreme gradient boosting (XGBoost) algorithm for accurate identification of type IV effectors based on optimal features based on protein sequences. After trying 20 different types of features, the best performance was achieved when all features were fed into XGBoost by the 5-fold cross validation in comparison with other machine learning methods. Then, the ReliefF algorithm was adopted to get the optimal feature set on our dataset, which further improved the model performance. T4SE-XGB exhibited highest predictive performance on the independent test set and outperformed other published prediction tools. Furthermore, the SHAP method was used to interpret the contribution of features to model predictions. The identification of key features can contribute to improved understanding of multifactorial contributors to host-pathogen interactions and bacterial pathogenesis. In addition to type IV effector prediction, we believe that the proposed framework can provide instructive guidance for similar studies to construct prediction methods on related biological problems. The data and source code of this study can be freely accessed at https://github.com/CT001002/T4SE-XGB.

Drug repositioning by prediction of drug's anatomical therapeutic chemical code via network-based inference approaches

Drug discovery and development is a time-consuming and costly process. Therefore, drug repositioning has become an effective approach to address the issues by identifying new therapeutic or pharmacological actions for existing drugs. The drug's anatomical therapeutic chemical (ATC) code is a hierarchical classification system categorized as five levels according to the organs or systems that drugs act and the pharmacology, therapeutic and chemical properties of drugs. The 2nd-, 3rd- and 4th-level ATC codes reserved the therapeutic and pharmacological information of drugs. With the hypothesis that drugs with similar structures or targets would possess similar ATC codes, we exploited a network-based approach to predict the 2nd-, 3rd- and 4th-level ATC codes by constructing substructure drug-ATC (SD-ATC), target drug-ATC (TD-ATC) and Substructure&Target drug-ATC (STD-ATC) networks. After 10-fold cross validation and two external validations, the STD-ATC models outperformed the SD-ATC and TD-ATC ones. Furthermore, with KR as fingerprint, the STD-ATC model was identified as the optimal model with AUC values at 0.899 ± 0.015, 0.916 and 0.893 for 10-fold cross validation, external validation set 1 and external validation set 2, respectively. To illustrate the predictive capability of the STD-ATC model with KR fingerprint, as a case study, we predicted 25 FDA-approved drugs (22 drugs were actually purchased) to have potential activities on heart failure using that model. Experiments in vitro confirmed that 8 of the 22 old drugs have shown mild to potent cardioprotective activities on both hypoxia model and oxygen-glucose deprivation model, which demonstrated that our STD-ATC prediction model would be an effective tool for drug repositioning.

STS-NLSP: A Network-Based Label Space Partition Method for Predicting the Specificity of Membrane Transporter Substrates Using a Hybrid Feature of Structural and Semantic Similarity

Membrane transport proteins play crucial roles in the pharmacokinetics of substrate drugs, the drug resistance in cancer and are vital to the process of drug discovery, development and anti-cancer therapeutics. However, experimental methods to profile a substrate drug against a panel of transporters to determine its specificity are labor intensive and time consuming. In this article, we aim to develop an in silico multi-label classification approach to predict whether a substrate can specifically recognize one of the 13 categories of drug transporters ranging from ATP-binding cassette to solute carrier families using both structural fingerprints and chemical ontologies information of substrates. The data-driven network-based label space partition (NLSP) method was utilized to construct the model based on a hybrid of similarity-based feature by the integration of 2D fingerprint and semantic similarity. This method builds predictors for each label cluster (possibly intersecting) detected by community detection algorithms and takes union of label sets for a compound as final prediction. NLSP lies into the ensembles of multi-label classifier category in multi-label learning field. We utilized Cramér's V statistics to quantify the label correlations and depicted them via a heatmap. The jackknife tests and iterative stratification based cross-validation method were adopted on a benchmark dataset to evaluate the prediction performance of the proposed models both in multi-label and label-wise manner. Compared with other powerful multi-label methods, ML-kNN, MTSVM, and RAkELd, our multi-label classification model of NLPS-RF (random forest-based NLSP) has proven to be a feasible and effective model, and performed satisfactorily in the predictive task of transporter-substrate specificity. The idea behind NLSP method is intriguing and the power of NLSP remains to be explored for the multi-label learning problems in bioinformatics. The benchmark dataset, intermediate results and python code which can fully reproduce our experiments and results are available at https://github.com/dqwei-lab/STS.