Inductive matrix completion for predicting gene–disease associations

By Nagarajan Natarajan, Inderjit S. Dhillon. Bioinform., 2014

315 citations recorded by the source corpus.

Motivation: Most existing methods for predicting causal disease genes rely on specific type of evidence, and are therefore limited in terms of applicability. More often than not, the type of evidence available for diseases varies—for example, we may know linked genes, keywords associated with the disease obtained by mining text, or co-occurrence of disease symptoms in patients. Similarly, the type of evidence available for genes varies—for example, specific microarray probes convey information only for certain sets of genes. In this article, we apply a novel matrix-completion method called Inductive Matrix Completion to the problem of predicting gene-disease associations; it combines multiple types of evidence (features) for diseases and genes to learn latent factors that explain the observed gene–disease associations. We construct features from different biological sources such as microarray expression data and disease-related textual data. A crucial advantage of the method is that it is inductive; it can be applied to diseases not seen at training time, unlike traditional matrix-completion approaches and network-based inference methods that are transductive. Results: Comparison with state-of-the-art methods on diseases from the Online Mendelian Inheritance in Man (OMIM) database shows that the proposed approach is substantially better—it has close to one-in-four chance of recovering a true association in the top 100 predictions, compared to the recently proposed Catapult method (second best) that has <15% chance. We demonstrate that the inductive method is particularly effective for a query disease with no previously known gene associations, and for predicting novel genes, i.e. genes that are previously not linked to diseases. Thus the method is capable of predicting novel genes even for well-characterized diseases. We also validate the novelty of predictions by evaluating the method on recently reported OMIM associations and on associations recently reported in the literature. Availability: Source code and datasets can be downloaded from http://bigdata.ices.utexas.edu/project/gene-disease. Contact: naga86@cs.utexas.edu

Subfields: Molecular Biology

View Paper: https://www.semanticscholar.org/paper/02e34d24ebab52ee516c2ab50d93d360bde68187

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

Citations in the corpus

Multitask Matrix Completion for Learning Protein Interactions Across Diseases (2016, methodology): “IMC (Jain and Dhillon, 2013; Natarajan and Dhillon, 2014): This is the link-prediction model from Section 2, in which data from all tasks are combined without incorporating any task relationships.”
NeoDTI: Neural integration of neighbor information from a heterogeneous network for discovering new drug-target interactions (2018, methodology): “DTINet [5] first uses an unsupervised manner to automatically learn the low-dimensional feature representations of drugs and targets from heterogeneous network data, and then applies inductive matrix completion [19] to predict new DTIs based on the learnt features.”
Network Representation Learning with Rich Text Information (2015, methodology): “(7) Note that, inductive matrix completion is originally proposed to complete gene-disease matrix with gene and disease features [Natarajan and Dhillon, 2014], the goal of which is quite different from that of our work.”
Deep Representation Learning for Social Network Analysis (2019, methodology): “Inspired by the idea of inductive matrix completion (Natarajan and Dhillon, 2014), TADW (Yang et al.”
A network integration approach for drug-target interaction prediction and computational drug repositioning from heterogeneous information (2017, methodology): “As in the previous works 25, 26 , we used “ recall @ top-k ” as the evaluation metric, which is de ﬁ ned as the fraction of true interacting targets that were retrieved in the list of top-k predictions for a drug.”
Predicting miRNA‐disease association based on inductive matrix completion (2018, background): “Several articles (Fantine and Jean-Philippe, 2011; Natarajan and Dhillon, 2014; Singh-Blom et al., 2013) have adopted this performance measure for evaluation.”
Network Representation Learning: A Survey (2017, methodology): “For example, in the Modularity Maximization method [31], eigen decomposition is performed on the modularity matrix to learn community indicative vertex representations [53]; in the TADW algorithm [7], inductive matrix factorization [54] is carried out on the vertexcontext matrix to simultaneously pr”
A Survey on Network Embedding (2017, methodology): “Some multimodal and multisource fusion techniques are explored in this line of research [15], [17].”