A Novel XGBoost Method to Infer the Primary Lesion of 20 Solid Tumor Types From Gene Expression Data

Abstract

Purpose: Establish a suitable machine learning model to identify its primary lesions for primary metastatic tumors in an integrated learning approach, making it more accurate to improve primary lesions’ diagnostic efficiency. Methods: After deleting the features whose expression level is lower than the threshold, we use two methods to perform feature selection and use XGBoost for classification. After the optimal model is selected through 10-fold cross-validation, it is verified on an independent test set. Results: Selecting features with around 800 genes for training, the R2-score of a 10-fold CV of training data can reach 96.38%, and the R2-score of test data can reach 83.3%. Conclusion: These findings suggest that by combining tumor data with machine learning methods, each cancer has its corresponding classification accuracy, which can be used to predict primary metastatic tumors’ location. The machine-learning-based method can be used as an orthogonal diagnostic method to judge the machine learning model processing and clinical actual pathological conditions.