Table 2:

Multiple alignment programme scores for the alignment reference sets in Balibase 3.0

MSA programme	Reference 1 Equidistant sequences		Reference 2 Family with orphans	Reference 3 Divergent sub families	Reference 4 Large extensions	Reference 5 Large insertions	Reference 7 Trans-membrane proteins	Reference 9 Linear motifs
	V1: <20%	V2: 20–40%
ProbCons	0.65	0.93	0.90	0.79	0.86	0.88	0.81	0.68
	36.8%	68.2%	34.1%	43.3%	63.4%	62.5%	50.0%	50.7%
MAFFT	0.45	0.88	0.88	0.74	0.83	0.79	0.81	0.66
	44.7%	20.4%	51.2%	46.7%	41.5%	31.2%	37.5%	22.4%
Muscle	0.56	0.90	0.88	0.76	0.84	0.83	0.79	0.66
	13.2%	4.5%	9.8%	10.0%	9.8%	6.2%	12.5%	4.5%
ClustalW	0.46	0.85	0.86	0.62	0.75	0.61	0.69	0.63
	5.3%	6.8%	4.9%	0.0%	4.9%	0.0%	0.0%	22.4%

For each programme, the mean SP scores are shown, as well as the number of times the programme achieved the best alignment score, expressed as a percentage of the total number of reference alignments. The programme with the highest percentage of best scores is highlighted in bold. References 6 and 8 were excluded, since the tests contained non-linear sequences that cannot be aligned by most multiple alignment methods.