EARN as a precision oncology tool leads us to propose the targeted genes panel for metastatic breast cancer

Today, there are a lot of bio-markers on the prognosis and diagnosis of complex diseases such as primarybreast cancer. However, our understanding of the drivers that influence cancer aggression is limited. Presentinvestigation studies somatic mutation data consisting of ۴۵۰ metastatic breast tumor samples from cBioCancer Genomics Portal. We use four software tools to extract features from this data. Then, an ensembleclassifier learning algorithm called EARN (Ensemble of Artificial Neural Network, Random Forest, and nonlinearSupport Vector Machine) is proposed to evaluate plausible driver genes for metastatic breast cancer(MBCA). It is an attempt to focus on the findings in four aspects of MBCA prognosis. First, drivers andpassengers predicted by SVM, ANN, RF, and EARN are introduced. Second, the performance of fourlearning methods is evaluated using statistical criteria. Third, the outputs of the biological inference based ongene set enrichment analysis (GSEA) and pathway enrichment analysis (PEA) are discussed. Finally, thePEA using ReactomeFIVIz tool (FDR<۰.۰۳) for the top ۱۰۰ predicted genes by EARN leads us to propose anew gene set panel for MBCA, including HDAC۳, ABAT, GRIN۱, PLCB۱, and KPNA۲ as well as NCOR۱,TBL۱XR۱, SIRT۴, KRAS, CACNA۱E, PRKCG, GPS۲, SIN۳A, ACTB, KDM۶B, and PRMT۱.Furthermore, we compare results for MBCA to other outputs regarding ۹۸۳ primary breast invasivecarcinoma (BRCA) tumor samples obtained from The Cancer Genome Atlas (TCGA). Meanwhile, the ۱۶-gene panel proposed by EARN has been surveyed in the whole-exome sequence of an archived FFPE sampleobtained from breast tissue of an anonymous Iranian female patient with invasive breast carcinoma. Thisresearch leverages both computational and experimental approaches to assist precision oncologists to designcompact targeted panels that eliminate the need for whole-genome/exome sequencing.

کلیدواژه ها:

Metastasis breast tumor ، Mutation data ، Ensemble classifier ، Plausible driver genes ، Targeted gene panel