Towards a better solution to the shortest common supersequence problem: the deposition and reduction algorithm

BMC Bioinformatics

Table 2 A comparison of the lengths of the SCS results obtained by different algorithms on simulated DNA sequences.

*Sequences*		*Length of CS (averaged over 10 instances)*
k	n	LB	*ALPHA*	*Ratio*	*TOUR*	*Ratio*	*GRDY*	*Ratio*	MM	*Ratio*	DR	*Ratio*
100	100	158.0	400	2.53	304.6 (15.3)	1.93	303.9 (11.9)	1.92	276.8 (5.5)	1.75	264.5 (6.8)	1.67
500	100	160.6	400	2.49	329.0 (18.2)	2.05	321.6 (9.1)	2.00	286.4 (7.7)	1.78	273.0 (7.1)	1.70
1000	100	161.3	400	2.48	335.4 (19.9)	2.08	329.0 (17.8)	2.04	289.1 (8.2)	1.79	275.6 (6.2)	1.71
5000	100	162.9	400	2.46	339.8 (21.2)	2.09	336.2 (21.2)	2.06	294.8 (10.6)	1.81	279.9 (4.2)	1.72
100	1000	1441.8	4000	2.77	2936.7 (146.6)	2.04	2921.5 (143.4)	2.03	2547.1 (24.7)	1.77	2480.2 (55.5)	1.72
500	1000	1457.6	4000	2.74	3049.6 (150.0)	2.09	3043.6 (145.2)	2.09	2578.0 (23.3)	1.77	2527.2 (52.1)	1.73
1000	1000	1472.3	4000	2.72	3142.3 (176.9)	2.13	3115.5 (173.6)	2.12	2590.9 (26.5)	1.76	2540.0 (47.4)	1.73
5000	1000	1481.6	4000	2.70	3194.5 (221.9)	2.16	3172.8 (199.0)	2.14	2602.1 (25.6)	1.76	2548.3 (39.7)	1.72

The average and standard deviation (in parenthesis) over 10 randomly generated instances are given. The estimated performance ratios are also given for each algorithm.

ISSN: 1471-2105