Heterogeneous network based data fusion can encode diverse inter- and intra-relations between objects, and has been sparking increasing attention in recent years. Matrix factorization based data fusion models have been invented to fuse multiple data sources. However, these models generally suffer from the widely-witnessed insufficient relations between nodes and from information loss when heterogeneous attributes of diverse network nodes are transformed into ad-hoc homologous networks for fusion. In this paper, we introduce a general data fusion model called Attributed Heterogeneous Network Fusion (AHNF). AHNF firstly constructs an attributed heterogeneous network composed with different types of nodes and the diverse attribute vectors of these nodes. It uses indicator matrices to differentiate the observed inter-relations from the latent ones, and thus reduces the impact of insufficient relations between nodes. Next, it collaboratively factorizes multiple adjacency matrices and attribute data matrices of the heterogeneous network into low-rank matrices to explore the latent relations between these nodes. In this way, both the network topology and diverse attributes of nodes are fused in a coordinated fashion. Finally, it uses the optimized low-rank matrices to approximate the target relational data matrix of objects and to effectively accomplish the relation prediction. We apply AHNF to predict the lncRNA-disease associations using diverse relational and attribute data sources. AHNF achieves a larger area under the receiver operating curve 0.9367 (by at least 2.14%), and a larger area under the precision-recall curve 0.5937 (by at least 28.53%) than competitive data fusion approaches. AHNF also outperforms competing methods on predicting de novo lncRNA-disease associations, and precisely identifies lncRNAs associated with breast, stomach, prostate, and pancreatic cancers. AHNF is a comprehensive data fusion framework for universal attributed multi-type relational data. The code and datasets are available at http://mlda.swu.edu.cn/codes.php?name=AHNF.

Matrix factorization based data fusion solutions can account for the intrinsic structures of multi-relational data sources, but most solutions equally treat these sources or prefer sparse ones, which may be irrelevant for the target task. In this paper, we introduce a Selective Matrix Factorization based Data Fusion approach (SelMFDF) to collaboratively factorize multiple inter-relational data matrices into low-rank representation matrices of respective object types and optimize the weights of them. To avoid preference to sparse data matrices, it additionally regularizes these low-rank matrices by approximating them to multiple intra-relational data matrices and also optimizes the weights of them. Both weights contribute to automatically integrate relevant data sources. Finally, it reconstructs the target relational data matrix using the optimized low-rank matrices. We applied SelMFDF for predicting inter-relations (lncRNA-miRNA interactions, functional annotations of proteins) and intra-relations (protein-protein interactions). SelMFDF achieves a higher AUROC (area under the receiver operating characteristics curve) by at least 5.88%, and larger AUPRC (area under the precision-recall curve) by at least 18.23% than other related and competitive approaches. The empirical study also confirms that SelMFDF can not only differentially integrate these relational data matrices, but also has no preference toward sparse ones.

Influx evidences show that red long non-coding RNAs (lncRNAs) play important roles in various critical biological processes, and they afffect the development and progression of various human diseases. Therefore, it is necessary to precisely identify the lncRNA-disease associations. The identification precision can be improved by developing data integrative models. However, current models mainly need to project heterogeneous data onto the homologous networks, and then merge these networks into a composite one for integrative prediction. We recognize that this projection overrides the individual structure of the heterogeneous data, and the combination is impacted by noisy networks. As a result, the performance is compromised. Given that, we introduce a weighted matrix factorization model on multi-relational data to predict LncRNA-disease associations (WMFLDA). WMFLDA firstly uses a heterogeneous network to capture the inter(intra)-associations between different types of nodes (including genes, lncRNAs, and Disease Ontology terms). Then, it presets weights to these inter-association and intra-association matrices of the network, and cooperatively decomposes these matrices into low-rank ones to explore the underlying relationships between nodes. Next, it jointly optimizes the low-rank matrices and the weights. After that, WMFLDA approximates the lncRNA-disease association matrix using the optimized matrices and weights, and thus to achieve the prediction. WMFLDA obtains a much better performance than related data integrative solutions across different experiment settings and evaluation metrics. It can not only respect the intrinsic structures of individual data sources, but can also fuse them with selection.

Background: Traditional drug research and development is high cost, time-consuming and risky. Computationally identifying new indications for existing drugs, referred as drug repositioning, greatly reduces the cost and attracts everincreasing research interests. Many network-based methods have been proposed for drug repositioning and most of them apply random walk on a heterogeneous network consisted with disease and drug nodes. However, these methods generally adopt the same walk-length for all nodes, and ignore the different contributions of different nodes. Results: In this study, we propose a drug repositioning approach based on individual bi-random walks (DR-IBRW) on the heterogeneous network. DR-IBRW firstly quantifies the individual work-length of random walks for each node based on the network topology and knowledge that similar drugs tend to be associated with similar diseases. To account for the inner structural difference of the heterogeneous network, it performs bi-random walks with the quantified walk-lengths, and thus to identify new indications for approved drugs. Empirical study on public datasets shows that DR-IBRW achieves a much better drug repositioning performance than other related competitive methods. Conclusions: Using individual random walk-lengths for different nodes of heterogeneous network indeed boosts the repositioning performance. DR-IBRW can be easily generalized to prioritize links between nodes of a network.

Increasing biomedical studies have demonstrated important associations between lncRNAs and various human complex diseases. Developing data integrative models can boost the performance of lncRNA-disease association identification. However, existing models generally have to transform heterogenous data into homologous networks, and then sum up these networks into a composite network for integrative prediction. The transformation may conceal the intrinsic structure of the heterogeneous data, and the summation process may suffer from noisy networks. Both these issues compromise the performance. In this paper, we introduce a Weighted Matrix Factorization based data fusion solution to predict LncRNA-Disease Associations (WMFLDA). WMFLDA first directly encodes the inter-associations between different types of biological entities (such as genes, lncRNAs, and Disease Ontology terms) via a heterogeneous network, which also encodes multiple intra-association networks of entities of the same type. Next, it assigns weights to these inter-association and intra-association matrices, and performs collaborative low-rank matrix factorization to explore the latent relationships between entities. After that, it simultaneously optimizes these weights and low-rank matrices. In the end, it uses the optimized low-rank matrices and weights to reconstruct the lncRNA-disease association matrix and accomplish the prediction. WMFLDA achieves a larger area under the receiver operating curve (by at least 7.61%), and a larger area under the precision-recall curve (by at least 5.49%) than competitive data fusion approaches in different experimental scenarios. WMFLDA can not only maintain the intrinsic structure of the association matrices, but can also selectively and differentially combine them. The codes and datasets are available at http: //mlda.swu.edu.cn/codes.php?name=WMFLDA.