Additional file 1 of Transductive learning as an alternative to translation initiation site identification

SVM parameters obtained by executing the Grid Search method Due to the high amount of available molecules (around 20 thousand) for the remaining analyzed organisms and the Grid Search’s high runtime (given by the SVM’s execution time and the amount of records in the training set), we use 10% of the available sequences. Those sequences were chosen using the Mersenne Twister method, but keeping the ratio of positive (TIS) and negative (nTIS) classes. Grid Search was executed for each of the organisms and window size defined in this work. This table presents the values for the parameters (C,γ) found by the Grid Search using RBF kernel function, which were used for the training of ISVM and TSVM. (XLS 28.0 kb)

通过网格搜索（Grid Search）方法获取得到的支持向量机（Support Vector Machine，SVM）参数。鉴于本次分析剩余受试生物的可用分子数量较多（约20000条），且网格搜索的运行时长较高（由支持向量机的执行时长与训练集记录数共同决定），我们选取了10%的可用序列。这些序列通过梅森旋转算法（Mersenne Twister）选取，且保留了阳性类别（TIS）与阴性类别（nTIS）的样本比例。本研究针对每种受试生物以及预设的窗口大小分别执行了网格搜索。本表格展示了通过径向基函数（Radial Basis Function，RBF）核函数的网格搜索得到的参数（C、γ）取值，这些参数被用于ISVM与TSVM的训练。（XLS 28.0 kb）