Not all transmembrane helices are born equal: Towards the extension of the sequence homology concept to membrane proteins

Table 1 Test of difference between the summary statistics of functional TM-helix sets (SCOP-derived versus UniProt-derived)

Section	Quantitative Criteria	Windows size	SCOP-derived			UniProt-derived
			μ _c	σ _c	n _c	μ _c	σ _c	n _c	p-value
A	Sequence complexity (based on IVL group)	10	2.31	0.28	83	2.28	0.29	1741	0.32
		12	2.42	0.29	83	2.40	0.30	1741	0.56
		15	2.56	0.31	83	2.53	0.32	1741	0.39
		18	2.68	0.31	83	2.63	0.32	1741	0.15
B	Hydrophobicity scale	19	0.41	2.91	83	0.64	2.85	1741	0.48

The summary statistics of the quantitative criteria (sequence complexity/hydrophobicity; column 2) for the SCOP-derived and UniProt-derived functional TM-helix sets are given in section A and B of Table 1 respectively (column 4-6, 7-9). The summary statistic (μ_c, σ_c, n_c) denotes the mean, standard deviation and sample size for the sequence complexity measures for a given window size (10,12,15 and 18; column 3) while the summary statistic (μ_Φ, σ_Φ, n_Φ) denotes the mean, standard deviation and sample size for the hydrophobicity measures for the window size of 19. The p-values (last column) are computed from the two-tailed t-test. None of the tests return a significant result which means that the summary statistics between the SCOP-derived and UniProt-derived functional TM-helix sets are similar. As such, calculations of false-negative rates based on either functional TM-helix sets should give similar conclusions.

ISSN: 1745-6150