Table 3.

Assessment of multiple sequence alignment programs using reference-independent sequence and structural similarity scores on 1207 representative SCOP40 domain pairs with identity <20%

Method	Structural similarity						Sequence similarity
	DALI Z-score	GDT-TS	TM-score	3D-score	LBcona	LBconb	Sequence identity	Blosum62 score
HMM_1_1_0	0.1178	0.2510	0.3005	0.2499a	0.2181	0.2828	0.0953	0.1687
HMM_1_1_1	0.1200a	0.2519a	0.3010a	0.2514a	0.2190a	0.2838	0.0955	0.1688
HMM_3_1_1	0.1217a	0.2540a	0.3034a	0.2532a	0.2215a	0.2872a	0.0938	0.1665
HMM_1_3_1	0.1226a	0.2564a	0.3061a	0.2557a	0.2230a	0.2892a	0.0944	0.1662
HMM_3_3_1	0.1231a	0.2570a	0.3070a	0.2563a	0.2240a	0.2909a	0.0932	0.1651
ProbCons	0.1003	0.2324	0.2767	0.2307	0.2060	0.2670	0.0983	0.1719
MAFFT-fftnsi	0.0982	0.2333	0.2814	0.2297	0.2004	0.2632	0.0917	0.1621
MAFFT-einsi	0.1136	0.2425	0.2886	0.2410	0.2105	0.2763	0.0940	0.1666
MAFFT-linsi	0.1135	0.2485	0.2982	0.2467	0.2143	0.2820	0.0923	0.1632
MAFFT-ginsi	0.1126	0.2454	0.2960	0.2429	0.2152	0.2803	0.0972	0.1725
MUSCLE	0.0980	0.2297	0.2777	0.2266	0.1941	0.2535	0.0939	0.1686
ClustalW	0.0723	0.1916	0.2318	0.1876	0.1551	0.2030	0.0733	0.1344

The first five methods are MUMMALS implementing different HMMs. The format of the HMM names (‘HMM_solv_ss_u’) is explained in Table 1. The best scores of MUMMALS and the best scores of other programs (ProbCons, MAFFT with different options, MUSCLE, ClustalW) are in bold.

^aMUMMALS with this model is statistically better than the best of other programs according to Wilcoxon signed-rank test (P < 0.01).