Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Features generated for computational splice-site prediction correspond to functional elements

Figure 1

Receiver Operating Curve Analysis for FGA, GeneSplicer and MaxEnt for Acceptor (A) and Donor (B) Splice-Site Prediction. The true positive rate (TP/(TP+FN)) is plotted versus the false positive rate (FP/(FP+TN)). We show the sensitivity values ranging from 50% to 95%. When the score threshold for each method is adjusted such that 5% of the true sites are missed (sensitivity is 95%), for acceptor splice-site prediction, MaxEnt has recalled 10.48 % of the false sites, GeneSplicer 5.80% and FGA only 2.49%, and, for donor splice-site prediction, MaxEnt has recalled 6.61 % of the false sites, GeneSplicer 6.40% and FGA only 3.30%. These results are computed on the Human dataset of GeneSplicer team which contains 1,115 pre-mRNA sequences.

Back to article page