The Neural Metric Factorization for Computational Drug Repositioning

Computational drug repositioning aims to discover new therapeutic diseases for marketed drugs and has the advantages of low cost, short development cycle, and high controllability compared to traditional drug development. The matrix factorization model has become the cornerstone technique for computational drug repositioning due to its ease of implementation and excellent scalability. However, the matrix factorization model uses the inner product operation to represent the association between drugs and diseases, which is lacking in expressive ability. Moreover, the degree of similarity of drugs or diseases could not be implied on their respective latent factor vectors, which is not satisfy the common sense of drug discovery. Therefore, a neural metric factorization model for computational drug repositioning (NMFDR) is proposed in this work. We novelly consider the latent factor vector of drugs and diseases as a point in the high-dimensional coordinate system and propose a generalized euclidean distance to represent the association between drugs and diseases to compensate for the shortcomings of the inner product operation. Furthermore, by embedding multiple drug (disease) metrics information into the encoding space of the latent factor vector, the information about the similarity between drugs (diseases) can be reflected in the distance between latent factor vectors. Finally, we conduct wide analysis experiments on three real datasets to demonstrate the effectiveness of the above improvement points and the superiority of the NMFDR model.

Automated summary

Computational drug repositioning aims to discover new therapeutic diseases for marketed drugs, providing advantages over traditional drug development in terms of cost, development cycle, and controllability. The matrix factorization model is widely used due to its easy implementation and scalability, but lacks expressive ability in representing the association between drugs and diseases. To overcome this, a neural metric factorization model (NMFDR) is proposed, considering the latent factor vectors of drugs and diseases as points in a high-dimensional coordinate system and introducing a generalized euclidean distance to represent their association. By embedding multiple drug (disease) metrics information into the latent factor vectors, the similarity between drugs (diseases) can be reflected in the distance between them. Experimental analysis on real datasets demonstrates the effectiveness and superiority of the NMFDR model.