A Non-EST-Based Method for Exon-Skipping Prediction

Table 1.

Features Differentiating Between Alternatively Spliced and Constitutively Spliced Exons




Alternatively spliced exons

Constitutively spliced exons

P-valuea
Average size 87 128 P < 10-16
Percent exons whose length is a multiple of 3 73% (177/243) 37% (642/1753) P < 10-9
Average human—mouse exon conservationb 94% 89% P < 10-36
Percent exons with upstream intronic elements conserved in mousec 92% (223/243) 45% (788/1753) P < 10-11
Percent exons with downstream intronic elements conserved in mousec 82% (199/243) 35% (611/1753) P <10-14
Percent exons with both upstream and downstream intronic elements conserved in mousec
77% (188/243)
17% (292/1753)
P < 10-37
  • a P-value was calculated using Fisher's exact test, except for the “average size” and “average human—mouse exon conservation”, for which P-value was calculated using student's t-test.

  • b Average percent of matching nucleotides in global alignment of the respective exons.

  • c The 100 intronic nucleotides immediately upstream (or downstream) of the exon were locally aligned with the mouse 100 counterpart intronic nucleotides using Sim4 (Florea et al. 1998). Conservation was defined if at least 12 consecutive perfectly matching nucleotides were found in the alignment.

This Article

  1. Genome Res. 14: 1617-1623

Preprint Server