Collaborative matrix factorization with multiple similarities for predicting drug-target interactions

By Xiaodong Zheng, Hao Ding, Hiroshi Mamitsuka, Shanfeng Zhu. Knowledge Discovery and Data Mining, 2013

328 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/92e8f21bcf8437b6e6b7e43d4c2e9bd717e41673

How researchers see this paper

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

In Artificial Intelligence

In Information Systems

In Statistical and Nonlinear Physics

Citations in the corpus

DeepDock: Enhancing Ligand-protein Interaction Prediction by a Combination of Ligand and Structure Information (2019, background): “In general, virtual screening has two main categories: ligand-based and structure-based [3].”
Machine Learning for Integrating Data in Biology and Medicine: Principles, Practice, and Opportunities (2018, background): “The learned latent space can be transformed into a drug-target network via a sequence of matrix operations and the resulting drug-target network is used to predict drug-target interactions [268].”
Neighborhood Regularized Logistic Matrix Factorization for Drug-Target Interaction Prediction (2016, methodology): “Therefore, NRLMF is more efficient than KBMF2K and CMF.”
A network integration approach for drug-target interaction prediction and computational drug repositioning from heterogeneous information (2017, methodology): “We compared DTINet with four state-of-the-art methods of DTI prediction methods, including BLMNII [39], NetLapRLS [76], HNM [70] and CMF [79].”
DeepConv-DTI: Prediction of drug-target interactions via deep learning with convolution on protein sequences (2018, methodology): “As another approach to DTI prediction models, matrix factorization methods have been recruited to predict DTIs, which approximate multiplying two latent matrices representing the compound and target protein to an interaction matrix and similarity score matrix [5, 6].”
MolTrans: Molecular Interaction Transformer for drug–target interaction prediction (2020, methodology): “Similarity-based methods such as kernel regression (Pahikkala et al., 2015) and matrix factorization (Zheng et al., 2013) methods exploit known DTI’s drug–target similarity information and infer new ones.”
Network-based machine learning and graph theory algorithms for precision oncology (2017, methodology): “Thepaper 71 proposed a collaborative matrix factorization method to factorize known drug – target relations to predict new relations constrained by the drug – drug similarity network and the target – target similarity network.”
NeoDTI: Neural integration of neighbor information from a heterogeneous network for discovering new drug-target interactions (2018, methodology): “Indeed, by integrating diverse information from heterogeneous data sources, methods like DTINet [5], MSCMF [17], HNM [18] can further improve the accuracy of DTI prediction.”
Large-Scale Off-Target Identification Using Fast and Accurate Dual Regularized One-Class Collaborative Filtering and Its Application to Drug Repurposing (2016, methodology): “…the best performed matrix factorization methods so far (NRLMF [17] and KBMF with twin kernels (KBMF2K) [16]), combination of WNN and GIP (WNNGIP [14]), and another type of matrix factorization method (Collaborative Matrix Factorization (CMF) [15]) for different types of chemicals and proteins.”
Machine learning approaches and databases for prediction of drug–target interaction: a survey paper (2020, methodology): “An approach to approximate the input DTI matrix by two low-rank matrices, which share the same feature space and are generated by the weighted similarity matrices of drugs and those of targets, respectively [182] using [183–186].”