Deep Learning Algorithms for Studying the Impact of Tumor Suppressor Gene Mutations on Breast Cancer

Breast cancer is the most common site of cancer causing death in women around the world. It is the most frequently diagnosed malignancy in women, and mutations in the tumor suppressor p53 are commonly detected in the most aggressive subtypes. Rapid advances in biological research over recent years have significantly enriched biological and medical data resources. Deep learning–based techniques have been successfully utilized to process data in this field, and they have exhibited state-of-the-art performances even on high-dimensional, nonstructural, and black-box biological data. In this article, we propose a deep learning approach by using recurrent neural networks to evaluate and assess the contribution of genetic mutations in the TP53 gene in the breast cancer. Moreover, preprocessing of the breast dataset (the genetic dataset used comprises TP53 gene sequences, for normal and breast cancer cases; 100 sequences of each class, obtained from NCBI, Ensembl, IGSR, and TCGA) was done by machine learning algorithms such as k-nearest neighbors and principal component analysis and artificial neural networks. The experimental results show that under a different dataset, the mutation on TP53 appears in about 80% of this dataset; accuracy achieved by the recurrent neural network model was 92%, and the precision was 91%. Finally, to enhance the performance and applicability of the model, it is recommended to focus on preprocessing stage and use different and cross-section modules.