Table 5.

Structural and sequential properties of loop-type specific words in Wset_≥30

Words characteristic	shared wordsa	short-long specific words	long-long specific words	flanking-region unspecific words	flanking-region specific words
Average RMSdw (Å)	0.90	0.80*	0.74	0.83	0.88
(± standard deviation)	(± 0.4)	(± 0.4)	(± 0.4)	(± 0.4)	(± 0.4)
AverageZmax	9.4	15*	9.7	8.5	11.6*
(± standard deviation)	(± 4.7)	(± 9.5)	(± 5.3)	(± 3.9)	(± 7.0)
Average nbpos*	3.0	4.6*	3.3	2.7	3.8
(± standard deviation)	(± 1.6)	(± 1.7)	(± 1.8)	(± 1.5)	(± 1.8)

The upper part of the table corresponds to the analysis of word structural properties. The intra-word structural variability is analysed using the Root Mean Square deviation (RMSd) between fragments corresponding to the same word (RMSd_w). The inter-word structural variability is analysed using the RMSd between fragments of two different words (RMSd_dev). The lower part of the table corresponds to the analysis of sequential properties of words. The intra-word amino-acid preferences of a word are analysed using Z_max criterion (cf. Method section) and the number of significant position of a word (nb_pos*). Numbers within brackets indicate standard deviations. *: significant differences according to the Kruskal-Wallis test. The RMSd_dev are computed on a subset of 890 words of Wset_≥30. ^a: words sharing by long and short loops