Somatic Copy Number Alteration-Based Prediction of Molecular Subtypes of Breast Cancer Using Deep Learning Model

Abstract

Statistical analysis of high throughput genomic data, such as gene expressions, copy number alterations (CNAs) and single nucleotide polymorphisms (SNPs), has become very popular in cancer studies in recent decades because such analysis can be very helpful to predict whether a patient has a certain cancer or its subtypes. However, due to the high-dimensional nature of the data sets with hundreds of thousands of variables and very small numbers of samples, traditional machine learning approaches, such as Support Vector Machines (SVMs) and Random Forests (RFs), cannot analyze these data efficiently. To overcome this issue, we propose a deep neural network model to predict molecular subtypes of breast cancer using somatic CNAs. Experiments show that our deep model works much better than traditional SVM and RF.