If you want to make sure all the functions running well, you have to manage your file uploaded as we suggest strictly.

The input of prediction is matrix with two or three columns (pre-miRNA identifier, pre-miRNA sequence, pre-miRNA secondary structure [not necessary]) separated with tab key, in which each line represents a pre-miRNA. An example of prediction dataset is shown in below:

The input for training is a matrix separated with tab key, in which each line represents a miRNA, and contains four or five description items (miRNA identifier, pre-miRNA identifier, miRNA sequence, pre-miRNA sequence and pre-miRNA secondary structure [not necessary]). The example for training dataset is shown in below:

If users would like to evaluate the prediction performance of miRLocator, one can upload the known miRNA datasets with the same format as training.