Protein network inference from multiple genomic data: a supervised approach

By Yoshihiro Yamanishi, Jean-Philippe Vert, Minoru Kanehisa. 2004

259 citations recorded by the source corpus.

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

Fast Learning from Distributed Datasets without Entity Matching (2016, background): “Fo instance, insurance and banking data together can improve fraud detection; shopping records wel l complement medical history for estimating risk of disease [29]; joining heterogeneous data h elps prediction in genomics [16, 33]; 2”
ProDiGe: Prioritization Of Disease Genes with multitask machine learning from positive and unlabeled examples (2011, methodology): “This simple approach is widely used and often leads to very good performance with SVM to learn classification models from heterogeneous information [22,30,31].”
Bioinformatics Original Paper Integrating Multi-attribute Similarity Networks for Robust Representation of the Protein Space (unknown): “The advantages of integrating multiple data sources for predicting protein function has been demonstrated by a number of recent studies (Deng et al., 2004; Lanckriet et al., 2004; Pavlidis et al., 2001; Yamanishi et al., 2004).”
Semi-supervised multi-task learning for predicting interactions between HIV-1 and human proteins (2010, unknown): “Availability: http://www.cs.cmu.edu/∼qyj/HIVsemi Contact: qyj@cs.cmu.edu”
Supervised reconstruction of biological networks with local models (2007, result): “Kernel metric learning (kML) (Vert and Yamanishi, 2005). A new metric between genes is learned before applying the direct approach with respect to the new metric. The new metric, which depends on the genomic data, is explicitly optimized to improve the agreement between the metric and the training g”
Semi-supervised Penalized Output Kernel Regression for Link Prediction (2011, unknown): “Several input features have already been used to infer this network: gene expressions, phylogenetic profiles, localization data and protein interaction data derived from yeast two-hybrid (Yamanishi et al., 2004; Kato et al., 2005; Geurts et al., 2006; Bleakley et al., 2007).”
Deciphering Protein–Protein Interactions. Part II. Computational Methods to Predict Protein and Domain Interaction Partners (2007, background): “Multiple sources of direct and indirect data on protein–protein interactions can be combined in a supervised learning framework or integrative scoring scheme to predict protein–protein and domain–domain interactions [”
Evaluation of different biological data and computational classification methods for use in protein interaction prediction (2006, methodology): “To predict pathway protein interactions, Yamanishi et al.(14) presented a method using a variant of kernel canonical correlation analysis applied to a detailed feature set.”
Input Output Kernel Regression (2016, methodology): “• kCCA (Yamanishi et al., 2004): kernel CCA is used to detect correlations existing between the input kernel and a di ↵ usion kernel derived from the adjacency matrix of the labeled PPI network.”
Inferring cellular networks – a review (2007, methodology): “[1] cross-validation can be applied to choose optimal parameters.”