Additional file 1 of Classification of tumor types using XGBoost machine learning model: a vector space transformation of genomic alterations

Additional file 1: Figure S1. Example of a SPM[t] dataset for a generic tumor type t. Figure S2. Example of a CNV[t] dataset for a generic tumor type t. Figure S3. Pseudocode of the VSM data transformation procedure. Figure S4. Charts showing the size, in terms of total count and percentage, of each random group in the newly created dataset with groups as targets and confusion matrix showing the performance [accuracy (ACC), balanced accuracy (BACC) and AUC score] of the model; hyperparameters are also reported. Of note, accuracy values obtained from random grouping experiments reported here, were significantly lower than those obtained by performing grouping experiments based on biological criteria and characterized by the same numerical complexity (similar group sizes).

附加文件1：图S1。通用肿瘤类型t的SPM[t]数据集示例。
附加文件1：图S2。通用肿瘤类型t的CNV[t]（拷贝数变异，Copy Number Variation）数据集示例。
附加文件1：图S3。VSM（向量空间模型，Vector Space Model）数据转换流程的伪代码。
附加文件1：图S4。展示了以分组为目标变量的新建数据集各随机组的总计数与占比规模，以及体现模型性能的混淆矩阵（含准确率（Accuracy, ACC）、平衡准确率（Balanced Accuracy, BACC）与AUC得分）；同时列出了所用超参数。值得注意的是，本研究中随机分组实验得到的准确率，显著低于组规模相近、数值复杂度一致的基于生物学标准的分组实验所得准确率。