Table 2.

Number of positions with a given G composition depending on context in the core region (803 positions) of bacterial and mmt SSU rRNA alignments

		Bacterial alignment				mmt alignment
G com- position	Structural Context	Number of core positions	Expected value	Percent of core positions in group	Percent of total core positions	Number of core positions	Expected value	Percent of core positions in group	Percent of total core positions
ALL	Helix	263	-	32.8%	32.8%	238	-	29.6%	29.6%
	Flanking	230	-	28.6%	28.6%	258	-	32.1%	32.1%
	Loop	310	-	38.6%	38.6%	307	-	38.2%	38.2%
	Total	803	-	100.0%	100.0%	803	-	100.0%	100.0%
≥95%	Helix	48	58.6	26.8%	6.0%	37	34.4	31.9%	4.6%
	Flanking	73	51.3	40.8%	9.1%	56	37.3	48.3%	7.0%
	Loop	58	69.1	32.4%	7.2%	23	44.3	19.8%	2.9%
	Group total	179	179.0	100.0%	22.3%	116	116.0	100.0%	14.4%
≥15% <95%	Helix	74	42.3	57.4%	9.2%	33	27.3	35.9%	4.1%
	Flanking	32	36.9	24.8%	4.0%	36	29.6	39.1%	4.5%
	Loop	23	49.8	17.8%	2.9%	23	35.2	25.0%	2.9%
	Group total	129	129.0	100.0%	16.1%	92	92.0	100.0%	11.5%
<15%	Helix	141	162.1	28.5%	17.6%	168	176.4	28.2%	20.9%
	Flanking	125	141.8	25.3%	15.6%	166	191.2	27.9%	20.7%
	Loop	229	191.1	46.3%	28.5%	261	227.5	43.9%	32.5%
	Group total	495	495.0	100.0%	61.6%	595	595.0	100.0%	74.1%

The structural context is classified as Helix (for those nts within a helix), Flanking (for those nts at the ends of helices), and Loop (for internal or hairpin loops). Using the simple null model that columns of the alignment correspond randomly and uniformly to positions in the RNA structure, the expected number of positions that would fall into each group was calculated using the following formula: expected number = (mumber of positions with the given G composition range) * (number of positions with the given structural context)/803. For example, the expected number of bacterial positions with ≥95%G and helix interior is (48+73+58)*(48+74+141)/803 = 58.62.