Thumbnails:
List:
Year:
Category:
Session:
Poster:
Getting poster data...
Bikash Ranjan Samal, Alagar S, Ranjit Prasad Bahadur (IIT Kharagpur, West Bengal, India, Pin code - 721302)
Several studies have been published with breast cancer gene regulatory networks, however the heterogeneous and complex classification of breast cancer phenotype is widely ignored while constructing these regulatory networks. Here in this study, we have devised a more accurate classification model for intrinsic subtypes of breast cancer based on their gene expression profile data using supervised machine learning, and reconstructed the gene regulatory network for each of the subtypes based on gene expression profile data by variable selection with ensembles of regression trees to infer links between genes. Regulatory networks are often incomplete without the multi-omics data like methylation, copy number abberation and miRNA expression profile data, hence we also proposed a model by integrating these multi-omics data to make these gene regulatory networks more accurate and meaningful. The reconstructed GRNs were analysed for their differential regulatory patterns with respect to their subtypes to identify unique features and crucial targets that will provide us a better level of description which will help in effective treatment and prognosis of breast cancer subtypes.