Multi-Objective Approach for Identifying Cancer Subnetwork Markers
Bayat, F.; Davoodi Monfared, M.
Show abstract
Identifying genetic markers for cancer is one of the main challenges in the recent researches. Between different cohorts of genetic markers such as genes or a group of genes like pathways or sub-network, identifying functional modules like subnetwork markers has been more challenging. Network-based classification methods have been successfully used for finding effective cancer subnetwork markers. Combination of metabolic networks and molecular profiles of tumor samples has led researchers to a more accurate prediction of subnetwork markers. However, topological features of the network have not been considered in the activity of the subnetwork. Here, we apply a novel protein-protein interaction network-based classification method that considers topological features of the network in addition to the expression profiles of the samples. We have considered the problem of identifying cancer subnetwork markers as a multi-objective problem which in this approach, each subnetworks activity level is measured according to both objectives of the problem; Differential expression level of the genes and topological features of the nodes in the network. We found that the subnetwork markers identified by this method achieve higher performance in the classification of cancer outcome in comparison to the other subnetwork markers.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Predicting Adverse Drug Effects: A Heterogeneous Graph Convolution Network with a Multi-layer Perceptron Approach 96%
- Improving prediction of drug-target interactions based on fusing multiple features with data balancing and feature selection techniques 96%
- Projection in genomic analysis: A theoretical basis to rationalize tensor decomposition and principal component analysis as feature selection tools 96%
Similar papers in this journal
- Comparing protein-protein interaction networks of SARS-CoV-2 and (H1N1) influenza using topological features 97%
- A Convolution Based Computational Approach Towards DNA N6-methyladenine Site Identification and Motif Extraction in Rice Genome 96%
- Classification models for Invasive Ductal Carcinoma Progression, based on gene expression data-trained supervised machine learning 96%
Similar papers in this journal
- Investigate the relevance of major signaling pathways in cancer survival using a biologically meaningful deep learning model 95%
- Combining Gene Ontology with Deep Neural Networks to Enhance the Clustering of Single Cell RNA-Seq Data 95%
- RAFTS3G - An efficient and versatile clustering software to analyses in large protein datasets 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.