Skip to main content

Table 3 Composition of the original G3PO and extended G3PO+ datasets

From: Spliceator: multi-species splice site prediction using convolutional neural networks

 

G3PO

Extension

G3PO+

Confirmed

889

472

1361

Unconfirmed

904

476

1380

Total

1793

948

2741

  1. To build the G3PO+ dataset, we retrieved orthologous sequences for 45 human genes and performed multiple sequence alignments. Each sequence was then checked to identify those that contained no errors, called ‘Confirmed’, and those that contained at least one error, called 'Unconfirmed'