Table 2. Open in new tab Results of...

Name	1A	1B	2A	2B	PPP score
Our results using software of Abeel et al. (2009)
RSV_Po2	0.18	0.28	0.42	0.55	0.30
RSV_RfG	0.18	0.28	0.41	0.56	0.30
RSV_Ex	0.18	0.28	0.41	0.56	0.30
Results in Abeel et al. (2009) with performance ≥ any RSV
ARTS	0.19	0.36	0.47	0.64	0.34
ProSOM	0.18	0.25	0.42	0.51	0.29
EP3	0.18	0.23	0.42	0.51	0.28
Eponine	0.14	0.29	0.41	0.57	0.27

Name	1A	1B	2A	2B	PPP score
Our results using software of Abeel et al. (2009)
RSV_Po2	0.18	0.28	0.42	0.55	0.30
RSV_RfG	0.18	0.28	0.41	0.56	0.30
RSV_Ex	0.18	0.28	0.41	0.56	0.30
Results in Abeel et al. (2009) with performance ≥ any RSV
ARTS	0.19	0.36	0.47	0.64	0.34
ProSOM	0.18	0.25	0.42	0.51	0.29
EP3	0.18	0.23	0.42	0.51	0.28
Eponine	0.14	0.29	0.41	0.57	0.27

Table 2.

Results of testing our predictors using the four benchmarks protocols of Abeel et al. (2009) and then comparing against 17 algorithms they evaluated.

Name	1A	1B	2A	2B	PPP score
Our results using software of Abeel et al. (2009)
RSV_Po2	0.18	0.28	0.42	0.55	0.30
RSV_RfG	0.18	0.28	0.41	0.56	0.30
RSV_Ex	0.18	0.28	0.41	0.56	0.30
Results in Abeel et al. (2009) with performance ≥ any RSV
ARTS	0.19	0.36	0.47	0.64	0.34
ProSOM	0.18	0.25	0.42	0.51	0.29
EP3	0.18	0.23	0.42	0.51	0.28
Eponine	0.14	0.29	0.41	0.57	0.27

Name	1A	1B	2A	2B	PPP score
Our results using software of Abeel et al. (2009)
RSV_Po2	0.18	0.28	0.42	0.55	0.30
RSV_RfG	0.18	0.28	0.41	0.56	0.30
RSV_Ex	0.18	0.28	0.41	0.56	0.30
Results in Abeel et al. (2009) with performance ≥ any RSV
ARTS	0.19	0.36	0.47	0.64	0.34
ProSOM	0.18	0.25	0.42	0.51	0.29
EP3	0.18	0.23	0.42	0.51	0.28
Eponine	0.14	0.29	0.41	0.57	0.27

We show the results for ProSOM (Abeel et al., 2008a), EP3 (Abeel et al., 2008b), Eponine (Down and Hubbard, 2002) and ARTS. The results are sorted according to the PPP score, which is the harmonic mean of the four individual scores giving an overall figure of merit. All results better than our ChIP-Seq developed model RSV_Po2 are marked in bold face; apart from ARTS there are only the two such results, both with the Eponine algorithm. The results for the remaining 13 algorithms have worse performance than every RSV for tests 1A–2B and can be found in Table 2Abeel et al. (2009).

This Feature Is Available To Subscribers Only