Identifying possible lncRNA-disease associations based on deep learning and positive-unlabeled learning

Conventional supervised binary classification algorithms have been widely applied to address significant research questions using biological and biomedical data. This classification scheme requires two fully labeled classes of data (e.g. positive and negative samples) to train a classification model. However, in many bioinformatics applications, labeling data is laborious, and the negative samples might be potentially mislabeled due to the limited sensitivity of the experimental equipment. The positive unlabeled (PU) learning scheme was therefore proposed to enable the classifier to learn directly from limited positive samples and a large number of unlabeled samples (i.e. a mixture of positive or negative samples). To date, several PU learning algorithms have been developed to address various biological questions, such as sequence identification, functional site characterization and interaction prediction. In this paper, we revisit a collection of 29 state-of-the-art PU learning bioinformatic applications to address various biological questions. Various important aspects are extensively discussed, including PU learning methodology, biological application, classifier design and evaluation strategy. We also comment on the existing issues of PU learning and offer our perspectives for the future development of PU learning applications. We anticipate that our work serves as an instrumental guideline for a better understanding of the PU learning framework in bioinformatics and further developing next-generation PU learning frameworks for critical biological applications.

Tax evasion detection has a crucial role in addressing tax revenue loss. In the real world, an accessed tax dataset only contains a small number of labeled taxpayers who evade tax (positive samples) and a large number of unlabeled taxpayers who either evade tax or do not evade tax. It is difficult to address this issue due to this nontraditional dataset. In addition, the basic features of taxpayers designed according to tax experts’ domain knowledge and experience are very limited to determining whether taxpayers evade tax. These limitations motivate the contribution of this work. In this paper, we argue that the tax evasion detection task in the real world should be formalized as a positive unlabeled (PU) learning problem. We propose a novel tax evasion detection method based on PU learning with Network Embedding features (PUNE). PUNE effectively detects tax evasion based on basic features and transaction network features that are extracted by a network embedding algorithm. Moreover, PUNE can work well even under label noise. To evaluate the effectiveness of PUNE, we conduct experimental tests on a real-world tax dataset. The results demonstrate that PUNE can significantly improve the performance of tax evasion detection.

lncRNAs are densely related to many human diseases. Identifying new lncRNA-disease associations (LDAs) conduces to better deciphering mechanisms of diseases, finding new biomarkers, and further promoting their diagnosis and treatment. In this manuscript, we devise an LDA prediction framework called LDA-GARB. LDA-GARB first combines nonnegative matrix factorization to extract linear features of lncRNAs and diseases. Next, it computes lncRNA similarity and disease similarity and adopts a graph autoencoder to extract nonlinear features of lncRNAs and diseases. Subsequently, the extracted features are concatenated as a vector. Finally, it takes the obtained vector as inputs and designs a noise-robust gradient boosting model to uncover potential associations from unknown lncRNA-disease pairs. To investigate the LDA-GARB performance, we used precision, recall, accuracy, F1-score, AUC, and AUPR as measurement metrics and performed multiple comparison experiments. First, it was benchmarked with four representative LDA prediction methods, i.e., SDLDA, LDNFSGB, LDAenDL, and LDA-VGHB, under 5-fold cross validations on lncRNAs, diseases, and lncRNA-disease pairs. Next, it was compared with four representative boosting models, i.e., XGBoost, AdaBoost, CatBoost, and LightGBM, under the above three different cross validations. Subsequently, the performance of LDA-GARB against LDA-LNSUBRW, GAMCLDA, LDA-VGHB, LDAGM, and GANLDA on imbalanced data was evaluated. We also performed parameter sensitivity analysis and ablation experiments. The results demonstrated that LDA-GARB improved LDA prediction. Finally, LDA-GARB was applied to predict potential associated lncRNAs for colorectal cancer and breast cancer. CCDC26 and HAR1A have been inferred to have an association with the two cancers, respectively. As a useful LDA identification tool, LDA-GARB is freely available at https://github.com/smiling199/LDA-GARB.

Positive-unlabeled learning for open set domain adaptation

Positive-unlabeled (PU) learning deals with the binary classification problem when only positive (P) and unlabeled (U) data are available, without negative (N) data. Existing PU methods perform well on the balanced dataset. However, in real applications such as financial fraud detection or medical diagnosis, data are always imbalanced. It remains unclear whether existing PU methods can perform well on imbalanced data. In this paper, we explore this problem and propose a general learning objective for PU learning targeting specially at imbalanced data. By this general learning objective, state-of-the-art PU methods based on optimizing a consistent risk can be adapted to conquer the imbalance. We theoretically show that in expectation, optimizing our learning objective is equivalent to learning a classifier on the oversampled balanced data with both P and N data available, and further provide an estimation error bound. Finally, experimental results validate the effectiveness of our proposal compared to state-of-the-art PU methods.

Positive-unlabeled (PU) learning deals with the circumstances where only a portion of positive instances are labeled, while the rest and all negative instances are unlabeled, and due to this confusion, the class prior can not be directly available. Existing PU learning methods usually estimate the class prior by training a nontraditional probabilistic classifier, which is prone to give an overestimation. Moreover, these methods learn the decision boundary by optimizing the minimum margin, which is not suitable in PU learning due to its sensitivity to label noise. In this paper, we enhance PU learning methods from the above two aspects. More specifically, we first explicitly learn a transformation from unlabeled data to positive data by entropy regularized optimal transport to achieve a much more precise estimation for class prior. Then we switch to optimizing the margin distribution, rather than the minimum margin, to obtain a label noise insensitive classifier. Extensive empirical studies on both synthetic and real-world data sets demonstrate the superiority of our proposed method.

Epilepsy is a chronic disorder of the brain. Intracranial electroencephalogram (iEEG) recorded from cortex is the most popular measurement for not only the diagnosis of epilepsy, but also the focus localization that is crucial for the surgery. In recent years, the machine learning methods have been rapidly developed and applied successfully to various real world problems. Given sufficient number of samples, the powerful deep learning methods can achieve high performance for epileptic focus localization. However, it is a challenging task to obtain large amount of labeled iEEG regarding focal/non-focal channels, since the annotations must be performed by multiple clinical experts through visual judgment on the long term iEEG signals. In order to reduce the necessary number of labeled training samples, we introduce the positive unlabeled (PU) learning method for classification of focal and non-focal epileptic iEEG signals. The proposed method enables us to learn a binary classifier by using small amount of labeled data containing only one class (i.e., focal signals) and unlabeled data containing two classes (i.e., focal and non-focal signals), which greatly reduces the workload of clinical experts for annotations. Experimental results on Bern dataset and iEEG recorded from Juntendo University Hospital demonstrate the effectiveness of our method.

BackgroundMicrobes have dense linkages with human diseases. Balanced microorganisms protect human body against physiological disorders while unbalanced ones may cause diseases. Thus, identification of potential associations between microbes and diseases can contribute to the diagnosis and therapy of various complex diseases. Biological experiments for microbe–disease association (MDA) prediction are expensive, time-consuming, and labor-intensive.MethodsWe developed a computational MDA prediction method called GPUDMDA by combining graph attention autoencoder, positive-unlabeled learning, and deep neural network. First, GPUDMDA computes disease similarity and microbe similarity matrices by integrating their functional similarity and Gaussian association profile kernel similarity, respectively. Next, it learns the feature representation of each microbe–disease pair using graph attention autoencoder based on the obtained disease similarity and microbe similarity matrices. Third, it selects a few reliable negative MDAs based on positive-unlabeled learning. Finally, it takes the learned MDA features and the selected negative MDAs as inputs and designed a deep neural network to predict potential MDAs.ResultsGPUDMDA was compared with four state-of-the-art MDA identification models (i.e., MNNMDA, GATMDA, LRLSHMDA, and NTSHMDA) on the HMDAD and Disbiome databases under five-fold cross validations on microbes, diseases, and microbe-disease pairs. Under the three five-fold cross validations, GPUDMDA computed the best AUCs of 0.7121, 0.9454, and 0.9501 on the HMDAD database and 0.8372, 0.8908, and 0.8948 on the Disbiome database, respectively, outperforming the other four MDA prediction methods. Asthma is the most common chronic respiratory condition and affects ~339 million people worldwide. Inflammatory bowel disease is a class of globally chronic intestinal disease widely existed in the gut and gastrointestinal tract and extraintestinal organs of patients. Particularly, inflammatory bowel disease severely affects the growth and development of children. We used the proposed GPUDMDA method and found that Enterobacter hormaechei had potential associations with both asthma and inflammatory bowel disease and need further biological experimental validation.ConclusionThe proposed GPUDMDA demonstrated the powerful MDA prediction ability. We anticipate that GPUDMDA helps screen the therapeutic clues for microbe-related diseases.

Accumulating studies have shown that long non-coding RNAs (lncRNAs) are involved in many biological processes and play important roles in a variety of complex human diseases. Developing effective computational models to identify potential relationships between lncRNAs and diseases can not only help us understand disease mechanisms at the lncRNA molecular level, but also promote the diagnosis, treatment, prognosis, and prevention of human diseases. For this paper, a network-based model called NBLDA was proposed to discover potential lncRNA–disease associations, in which two novel lncRNA–disease weighted networks were constructed. They were first based on known lncRNA–disease associations and topological similarity of the lncRNA–disease association network, and then an lncRNA–lncRNA weighted matrix and a disease–disease weighted matrix were obtained based on a resource allocation strategy of unequal allocation and unbiased consistence. Finally, a label propagation algorithm was applied to predict associated lncRNAs for the investigated diseases. Moreover, in order to estimate the prediction performance of NBLDA, the framework of leave-one-out cross validation (LOOCV) was implemented on NBLDA, and simulation results showed that NBLDA can achieve reliable areas under the ROC curve (AUCs) of 0.8846, 0.8273, and 0.8075 in three known lncRNA–disease association datasets downloaded from the lncRNADisease database, respectively. Furthermore, in case studies of lung cancer, leukemia, and colorectal cancer, simulation results demonstrated that NBLDA can be a powerful tool for identifying potential lncRNA–disease associations as well.

BackgroundAccumulated evidence shows that the abnormal regulation of long non-coding RNA (lncRNA) is associated with various human diseases. Accurately identifying disease-associated lncRNAs is helpful to study the mechanism of lncRNAs in diseases and explore new therapies of diseases. Many lncRNA-disease association (LDA) prediction models have been implemented by integrating multiple kinds of data resources. However, most of the existing models ignore the interference of noisy and redundancy information among these data resources.ResultsTo improve the ability of LDA prediction models, we implemented a random forest and feature selection based LDA prediction model (RFLDA in short). First, the RFLDA integrates the experiment-supported miRNA-disease associations (MDAs) and LDAs, the disease semantic similarity (DSS), the lncRNA functional similarity (LFS) and the lncRNA-miRNA interactions (LMI) as input features. Then, the RFLDA chooses the most useful features to train prediction model by feature selection based on the random forest variable importance score that takes into account not only the effect of individual feature on prediction results but also the joint effects of multiple features on prediction results. Finally, a random forest regression model is trained to score potential lncRNA-disease associations. In terms of the area under the receiver operating characteristic curve (AUC) of 0.976 and the area under the precision-recall curve (AUPR) of 0.779 under 5-fold cross-validation, the performance of the RFLDA is better than several state-of-the-art LDA prediction models. Moreover, case studies on three cancers demonstrate that 43 of the 45 lncRNAs predicted by the RFLDA are validated by experimental data, and the other two predicted lncRNAs are supported by other LDA prediction models.ConclusionsCross-validation and case studies indicate that the RFLDA has excellent ability to identify potential disease-associated lncRNAs.

Similarity-guided graph contrastive learning for lncRNA-disease association prediction

Accumulated evidence indicates that lncRNAs are critical for many biological processes, especially diseases. Therefore, identifying potential lncRNA-disease associations is significant for disease prevention, diagnosis, treatment and understanding of cell life activities at the molecular level. Although novel technologies have generated considerable associations for various lncRNAs and diseases, it has inevitable drawbacks such as high cost, time consumption, and error rate. For this reason, integrating various biological databases to predict the potential association of lncRNA and disease is of great attraction. In this paper, we proposed the model called ECLDA to predict lncRNA-disease associations by combining CNN and highspeed ELM. Firstly, the feature vectors are constructed by integrating lncRNA functional similarity, disease semantic similarity and Gaussian interaction profile kernel similarity. Secondly, CNN is carried out to mine local and higher-level abstract features of the vectors. Finally, high speed ELM is used to identify the novel lncRNA disease associations. The ECLDA computational model achieved AUCs of 0.9014 in 5-fold cross validation. The results showed that ECLDA is expected to be a practical tool for biomedical research in the future.

It has been proved that long noncoding RNA (lncRNA) plays critical roles in many human diseases. Therefore, inferring associations between lncRNAs and diseases can contribute to disease diagnosis, prognosis and treatment. To overcome the limitation of traditional experimental methods such as expensive and time-consuming, several computational methods have been proposed to predict lncRNA-disease associations by fusing different biological data. However, the prediction performance of lncRNA-disease associations identification needs to be improved. In this study, we propose a computational model (named LDICDL) to identify lncRNA-disease associations based on collaborative deep learning. It uses an automatic encoder to denoise multiple lncRNA feature information and multiple disease feature information, respectively. Then, the matrix decomposition algorithm is employed to predict the potential lncRNA-disease associations. In addition, to overcome the limitation of matrix decomposition, the hybrid model is developed to predict associations between new lncRNA (or disease) and diseases (or lncRNA). The ten-fold cross validation and de novo test are applied to evaluate the performance of method. The experimental results show LDICDL outperforms than other state-of-the-art methods in prediction performance.

C-PUGP: A cluster-based positive unlabeled learning method for disease gene prediction and prioritization

Deceptive reviews detection has attracted significant attention from both business and research communities. However, due to the difficulty of human labeling needed for supervised learning, the problem remains to be highly challenging. This paper proposed a novel angle to the problem by modeling PU (positive unlabeled) learning. A semi-supervised model, called mixing population and individual property PU learning (MPIPUL), is proposed. Firstly, some reliable negative examples are identified from the unlabeled dataset. Secondly, some representative positive examples and negative examples are generated based on LDA (Latent Dirichlet Allocation). Thirdly, for the remaining unlabeled examples (we call them spy examples), which can not be explicitly identified as positive and negative, two similarity weights are assigned, by which the probability of a spy example belonging to the positive class and the negative class are displayed. Finally, spy examples and their similarity weights are incorporated into SVM (Support Vector Machine) to build an accurate classifier. Experiments on gold-standard dataset demonstrate the effectiveness of MPIPUL which outperforms the state-of-the-art baselines.