A statistical approach for 5′ splice site prediction using short sequence motifs and without encoding sequence data

Table 1 Threshold values and estimates of AUC-ROC for the proposed approach under different window sizes

Different window sizes	7 bp	8 bp	9 bp	10 bp	11 bp
Different window sizes	(−3 ~ +4)	(−2 ~ +6)	(−3 ~ +6)	(−3 ~ +7)	(−4 ~ +7)
Number of TSS	770	1370	1960	2342	2530
Number of FSS	12805	33840	59097	74140	88346
Threshold ( ε )	[−4.00]	[−5.63]	[−7.16]	[−8.83]	[−9.97]
AUC-ROC	92.74	93.37	94.43	94.39	93.87
SE	0.006	0.004	0.003	0.003	0.003

(−) indicates from exon side and (+) indicate from intron side excluding GT. Values in the square brackets are the final threshold values.

ISSN: 1471-2105