We found that nearly all differences in Sn and Sp are highly significant between TSEBRA and the other methods (see Additional file 1 : Table S5), using McNemar’s [ 35 ] test and the chi-square test of homogeneity [ 36 ] The number of transcripts per gene selected by TSEBRA was on average 1.07 which is at the same level as BRAKER2, and lower than the average of 1.20 observed for BRAKER1.
1
—
—