The evolution, impact and properties of exonic splice enhancers

Table 4 Distribution of ESEs between introns, exon core and exon flanks

Dataset	Intron density median	Exon density median	Intron density median per hexamer	Exon density median per hexamer	P ^a	Ends density median	Core density median	Ends density median per hexamer	Core density median per hexamer	P ^b
ESR	0.453	0.533	0.0016	0.0019	2.4e-56	0.541	0.524	0.0019	0.0018	6.96e-11
INT2	0.310	0.447	0.001	0.0014	3.0e-60	0.460	0.437	0.0015	0.0014	1.8e-15
INT2.400	0.201	0.305	0.001	0.0016	3.8e-61	0.327	0.298	0.0017	0.0016	1.06e-16
INT3	0.106	0.171	0.0016	0.002	2.29e-61	0.187	0.165	0.0023	0.0019	1.59e-17
INT3.400	0.069	0.117	0.0013	0.002	2.7e-62	0.131	0.113	0.0024	0.0021	1.2e-18
Ke-ESE	0.503	0.701	0.0004	0.0006	2.1e-62	0.697	0.706	0.0006	0.0006	5.91e-20
Ke-ESE400	0.161	0.308	0.0004	0.0008	1.03e-64	0.307	0.309	0.0007	0.0008	0.0006
PESE	0.256	0.383	0.0011	0.0016	2.45e-59	0.385	0.378	0.0016	0.0016	0.0003
RESCUE	0.217	0.286	0.0009	0.0012	1.4e-60	0.326	0.281	0.0014	0.0012	7.0e-21

Analysis of the ESE density in introns and in exon cores and flanks, where introns are defined by the terminal 100 bases (200 bp in total) and exons flanks are the 50 bp at either end and 100 bp in the centre of the exon.
^aP value for Mann Whitney U test comparing ESEs between intron flanks and exon (50 bp first and last bases and 100 in the middle).
^bP value for the Mann Whitney U test comparing exon flanks and exon cores. We also provide data on hexamer density per hexamer to enable visual comparison between different datasets but these were not employed for any statistics. P values in bold are significant after Bonferonni correction, with P <0.05/9.

ISSN: 1474-760X