Correlation of Prognostic Factors of Invasive Lobular Carcinoma and Invasive Ductal Carcinoma

Abstract

One of the biggest risks facing women in the twenty-first century is breast cancer. Invasion lobular carcinoma and invasion ductal carcinoma are the two main categories into which it is divided. Omics data is used to identify predictive biomarker signatures for clinical applications to detect breast cancer. Predictive performance has significantly improved because of recent advancements in machine learning techniques. Here, we are using an approach built on symbolic regression called the QLattice on a variety of clinical omics data sets. Through the identification of potential regulatory interactions between biomolecules, this method creates efficient, high-performing models that can forecast and explain the results of a specific omics experiment. The models have the potential to make it easier to find new biomarker signatures due to their clarity and obvious functional shape, which make them simple and easy to comprehend. A comprehensive experimental investigation was conducted to assess the machine learning model's efficacy in terms of the Area under the Curve (AUC) for breast cancer. The outcomes, which were contrasted with other approaches, demonstrate the suggested framework's efficacy and capacity to beat the alternative algorithms in terms of AUC, which is 0.66. Here, we profiled breast tumors in detail, including ductal carcinoma, mixed carcinoma, and invasive lobular carcinoma, by using the Gaussian method and TNXB gene.