Qualitative classification of the breast cancer genome and clustering of the cancer gene network

Abstract

The purpose of cancer genome project is to classify the genetic variations that are related to clinical phenotypes. However, some studies showed that some specific cellular pathways are targeted by the cancer mutations genes. But a few of the pathway genes are mutated in each patient. In most approaches, only the existing pathways are considered and the topology of the pathways are ignored. Consequently, new attempts have been targeted on classifying significantly mutated subnetworks and combining them with cancer survival. We had proposed a novel bioinformatics pipeline to identify quantitative classification of the breast cancer genome to verify if the steps will be working or not on real dataset. We have generated a mutation matrix from the collected dataset and calculated pairwise gene similarity. After that, we have also done clustering of the identified cancer gene network, which may help cancer patients by suggesting optimal treatments. We hope our pipeline can also be used for other types of mutation data analysis.