A replica exchange Monte Carlo algorithm for protein folding in the HP model

BMC Bioinformatics

Table 2 Results on 2D benchmark sequences

ID	E*	${PERM}_{t_{e x p}}$	ACO-HPPFP-3	REMC_vshd	REMC_pm	REMC_m
S1-1	-9	-9 (< 1 sec)	-9 (< 1 sec)	-9 (< 1 sec)	-9 (< 1 sec)	-9 (< 1 sec)
S1-2	-9	-9 (< 1 sec)	-9 (< 1 sec)	-9 (< 1 sec)	-9 (< 1 sec)	-9 (< 1 sec)
S1-3	-8	-8 (2 sec)	-8 (2 sec)	-8 (< 1 sec)	-8 (< 1 sec)	-8 (< 1 sec)
S1-4	-14	-14 (< 1 sec)	-14 (4 sec)	-14 (15 sec)	-14 (< 1 sec)	-14 (< 1 sec)
S1-5	-23	-23 (2 sec)	-23 (1 min)	-23 (91% of runs 18 min)	-23 (< 1 sec)	-23 (< 1 sec)
S1-6	-21	-21 (3 sec)	-21 (15 sec)	-21 (98% of runs 19 min)	-21 (< 1 sec)	-21 (< 1 sec)
S1-7	-36	-36 (4 sec)	-36 (20 min)	-34 (33% of runs 33 min)	-36 (10 sec)	-36 (13 sec)
S1-8	-42	-42 (78 hrs)	-42 (1.5 hrs)	-35 (11% of runs 40 min)	-42 (5 sec)	-42 (6 sec)
S1-9	-53	-53 (1 min)	-53 (20% of runs 1 day)	-50 (5% of runs 19 min)	-53 (2 min)	-53 (38 sec)
S1-10	-50	-50 (20 min)	-49 (12 hrs)	-46 (5% of runs 41 min)	-50 (3.5 min)	-50 (8 min)
S1-11	-48	-48 (8 min)	-47 (10 hrs)	-46 (5% of runs 97 min)	-48 (1 min)	-48 (1.2 min)

Details on runs can be found in the text. Results for PERM and ACO-HPPFP-3 are reproduced from [9]. In all instances, REMC_vshdreports the worst running times followed by ACO-HPPFP-3. REMC_moutperforms PERM in 10 of 11 instances and REMC_pmreports better times than PERM in 9 of 11 instances. Details of the experimental protocol can be found in the text.

ISSN: 1471-2105