Title: A Feature Weighting-Assisted Approach for Cancer Subtypes Identification from Paired Expression profiles
Journal: IEEE/ACM Transactions on Computational Biology and Bioinformatics
Authors: Madhumita and Sushmita Paul
Brief overview
Identification of cancer subtypes is critically important for understanding the heterogeneity present in tumors. Integrating information from multiple sources, homogeneous groups for cancer can be identified. However, there is a lack of computational approaches to identify histological subtypes among the patients suffering from different types of cancers. In this regard, a robust regression based algorithm is developed by judiciously integrating multi-modal data. The multi-modal data are integrated using Similarity Network Fusion method. Later, the cancer subtypes are discovered using spectral clustering. The algorithm is performing better than other existing methods by manifold. The biological analysis of the obtained biomarkers revealed that they could be used to develop a potential diagnostic and prognostic tool.
Full article
#multimodality #multiomics #spectralclustering #similarity network fusion #featureweighting #cancerresearch #personalisedmedicine
