SCCPMD: Probability matrix decomposition method subject to corrected similarity constraints for inferring long non-coding RNA–disease associations

Abstract

Accumulating evidence has demonstrated various associations of long non-coding RNAs (lncRNAs) with human diseases, such as abnormal expression due to microbial influences that cause disease. Gaining a deeper understanding of lncRNA–disease associations is essential for disease diagnosis, treatment, and prevention. In recent years, many matrix decomposition methods have also been used to predict potential lncRNA-disease associations. However, these methods do not consider the use of microbe-disease association information to enrich disease similarity, and also do not make more use of similarity information in the decomposition process. To address these issues, we here propose a correction-based similarity-constrained probability matrix decomposition method (SCCPMD) to predict lncRNA–disease associations. The microbe-disease associations are first used to enrich the disease semantic similarity matrix, and then the logistic function is used to correct the lncRNA and disease similarity matrix, and then these two corrected similarity matrices are added to the probability matrix decomposition as constraints to finally predict the potential lncRNA–disease associations. The experimental results show that SCCPMD outperforms the five advanced comparison algorithms. In addition, SCCPMD demonstrated excellent prediction performance in a case study for breast cancer, lung cancer, and renal cell carcinoma, with prediction accuracy reaching 80, 100, and 100%, respectively. Therefore, SCCPMD shows excellent predictive performance in identifying unknown lncRNA–disease associations.

HPTRMF: Collaborative Matrix Factorization-Based Prediction Method for LncRNA-Disease Associations Using High-Order Perturbation and Flexible Trifactor Regularization

“…LDCMFC: 23 Matrix factorization method based on correlation entropy. SCCPMD: 22 Algorithm based on probabilistic matrix factorization and logistic function-corrected similarity. CKA-HGRTMF: 21 Matrix factorization method based on hypergraph regularization and the use of central kernel alignment algorithm to fuse multiple similar matrices.…”

“…Finally, they calculated the final score prediction matrix based on trifactor matrix factorization methods. Lin et al 22 emphasized the importance of relevant data in similarity matrices and simultaneously weakened the impact of irrelevant data on the factorization process. Therefore, they proposed the probability matrix factorization method, SCCPMD, preprocessing the similarity matrix using a logistic function, and then calculating the final score matrix based on probability matrix factorization methods.…”

Predicting lncRNA–Disease Associations Based on a Dual-Path Feature Extraction Network with Multiple Sources of Information Integration

“…GCHIRFLDA 12 is a method for extracting potential features using an autoencoder and combining it with a Random Forest classifier for prediction. SCCPMD 13 first performed similarity enhancement using logistic functions and then performs prediction of potential association pairs using the probabilistic matrix decomposition method with corrected similarity constraints. These traditional machine learning and matrix factorization methods have achieved acceptable performance, but they neglect the information regarding neighboring nodes that is available in both the LDA association network and the lncRNA−disease similarity network.…”