Identifying biomarkers of nottingham prognosis index in breast cancer survivability

Nottingham Prognostics Index (NPI) is a widely-used prognostics measure used to predict survival of operable primary breast cancer. The NPI value is calculated based on the size of the tumor, the number of lymph nodes and the grade of the tumor. This work builds a prediction model for the NPI NPI ≥ 3.4, where this threshold is the cut-off between high survival rate versus the low survival rate. In this study, we present a supervised learning method used to predict the breast cancer NPI. The objectives of this research are (i) build a diagnosis system for breast cancer NPI based on multi-omics data; (ii) find gene biomarkers for each low and high NPI scores; (iii) build a novel prediction model based on t-distributed stochastic neighbor embedding (t-SNE) and residual neural network (ResNet) to integrate multi-omics data in the classification mechanism. The results show that two sets of biomarkers that include two different omics, namely gene expression and copy number alteration, can be integrated in the model to achieve a high prediction accuracy. Findings in the literature confirm the associations between some of these genes and breast cancer.