BackgroundThe ability to predict the interaction of drugs with target proteins is essential to research and development of drug. However, the traditional experimental paradigm is costly, and previous in silico prediction paradigms have been impeded by the wide range of data platforms and data scarcity.ResultsIn this paper, we modeled the prediction of drug-target interactions as a binary classification task. Using transcriptome data from the L1000 database of the LINCS project, we developed a framework based on a deep-learning algorithm to predict potential drug target interactions. Once fully trained, the model achieved over 98% training accuracy. The results of our research demonstrated that our framework could discover more reliable DTIs than found by other methods. This conclusion was validated further across platforms with a high percentage of overlapping interactions.ConclusionsOur model’s capacity of integrating transcriptome data from drugs and genes strongly suggests the strength of its potential for DTI prediction, thereby improving the drug discovery process.

Drug repositioning strategies have improved substantially in recent years. At present, two advances are poised to facilitate new strategies. First, the LINCS project can provide rich transcriptome data that reflect the responses of cells upon exposure to various drugs. Second, machine learning algorithms have been applied successfully in biomedical research. In this paper, we developed a systematic method to discover novel indications for existing drugs by approaching drug repositioning as a multi-label classification task and used a Softmax regression model to predict previously unrecognized therapeutic properties of drugs based on LINCS transcriptome data. This approach to complete the said task has not been achieved in previous studies. By performing in silico comparison, we demonstrated that the proposed Softmax method showed markedly superior performance over those of other methods. Once fully trained, the method showed a training accuracy exceeding 80% and a validation accuracy of approximately 70%. We generated a highly credible set of 98 drugs with high potential to be repositioned for novel therapeutic purposes. Our case studies included zonisamide and brinzolamide, which were originally developed to treat indications of the nervous system and sensory organs, respectively. Both drugs were repurposed to the cardiovascular category.

No abstract

With the rapid development of high-throughput technologies, parallel acquisition of large-scale drug-informatics data provides huge opportunities to improve pharmaceutical research and development. One significant application is the purpose prediction of small molecule compounds, aiming to specify therapeutic properties of extensive purpose-unknown compounds and to repurpose novel therapeutic properties of FDAapproved drugs. Such problem is very challenging since compound attributes contain heterogeneous data with various feature patterns such as drug fingerprint, drug physicochemical property, drug perturbation gene expression. Moreover, there is complex nonlinear dependency among heterogeneous data. In this paper, we propose a novel domain-adversarial multitask framework for integrating shared knowledge from multiple domains. The framework utilizes the adversarial strategy to effectively learn target representations and models their nonlinear dependency. Experiments on two real-world datasets illustrate that the performance of our approach obtains an obvious improvement over competitive baselines. The novel therapeutic properties of purpose-unknown compounds we predicted are mostly reported or brought to the clinics. Furthermore, our framework can integrate various attributes beyond the three domains examined here and can be applied in the industry for screening the purpose of huge amounts of as yet unidentified compounds.

The standard algorithms of decision trees and their derived methods are usually constructed on the basis of the frequency information. However, they still suffer from a dilemma or multichotomous question for continuous attributes when two or more candidate cut points have the same or similar splitting performance with the optimal value, such as the maximal information gain ratio or the minimal Gini index. In this paper, we propose a unified framework model to deal with this question. We then design two algorithms based on Splitting Performance and the number of Expected Segments, called SPES1 and SPES2, which determine the optimal cut point, as follows. First, several candidate cut points are selected based on their splitting performances being the closest to the optimal. Second, we compute the number of expected segments for each candidate cut point. Finally, we combine these two measures by introducing a weighting factor α to determine the optimal one from several candidate cut points. To validate the effectiveness of our methods, we perform them on 25 benchmark datasets. The experimental results demonstrate that the classification accuracies of the proposed algorithms are superior to the current state-of-the-art methods in tackling the multichotomous question, about 5% in some cases. In particular, according to the proposed methods, the number of candidate cut points converges to a certain extent. INDEX TERMS Decision tree, classification, unified framework, split criteria.