# [AMBER] Two dimensional replica exchange MD: Success rate

From: Stephan Eismann <seismann.stanford.edu>
Date: Wed, 27 Feb 2019 17:30:20 -0800

Dear all,

I am running two dimensional replica exchange MD with Amber18. My question
relates to the log files produced and specifically to the column "success
rate".

I understand that success rate is calculated as (#successes / #tries) x 2
and #tries is indicated by the exchange number.

1) I am unable to reproduce "success rate" by tracking "successes" (T or
F) and #exchange based on the formula above. If somebody could explain me
how this is done for one example, I would greatly appreciate it.

2) Sometimes Rep# and Neibr# are identical. What does this mean and why do
the potential energies differ?

Please find below a simple example (the head of log file for the
Hamiltonian dimension) with 2 temperatures and 4 Hamiltonian modifications
(different boost factors for the dihedral energies).

Thank you,
Stephan

# Replica Exchange log file
# numexchg is 1000
# Dimension 2 of 2
# Description: Hamiltonian exchange
# exchange_type = HAMILTONIAN
# REMD filenames:
# remlog= rem.log.1.2
# remd dimension file= remd.dim
# Rep#, Neibr#, Temp0, PotE(x_1), PotE(x_2), left_fe, right_fe, Success,
Success rate (i,i+1)
# exchange 2 REMD group 2
1 2 310.50-131101.56-131419.90 0.00 0.03 T
2.00
2 1 310.50-131419.92-131101.59 -0.02 0.00 T
0.00
3 4 310.50-130997.50-130631.42 0.00 0.03 T
2.00
4 3 310.50-130631.43-130997.53 -0.01 0.00 T
0.00
# exchange 2 REMD group 1
1 2 310.00-131160.34-131176.52 0.00 0.02 T
2.00
2 1 310.00-131176.55-131160.36 -0.03 0.00 T
0.00
3 4 310.00-131115.67-131441.72 0.00 0.02 T
2.00
4 3 310.00-131441.75-131115.69 -0.03 0.00 T
0.00
# exchange 4 REMD group 2
1 2 310.50-130839.62-130967.64 0.02 0.03 F
1.00
4 1 310.50-130967.66-130839.64 -0.01 -0.02 F
0.00
2 4 310.50-130759.48-130989.12 -0.02 0.01 T
1.00
3 3 310.50-130989.13-130759.49 -0.01 0.03 T
1.00
# exchange 4 REMD group 1
2 2 310.00-131194.05-131349.69 -0.03 0.01 T
1.00
3 1 310.00-131349.71-131194.07 -0.02 0.02 T
0.00
1 4 310.00-130940.61-130929.28 0.01 0.02 T
0.00
4 3 310.00-130929.32-130940.62 -0.03 -0.04 T
1.00
# exchange 6 REMD group 1
1 2 310.00-130901.98-131287.82 0.01 0.01 T
1.33
2 1 310.00-131287.83-130901.99 -0.02 0.01 T
0.67
3 4 310.00-130774.52-131207.09 -0.02 0.01 T
1.33
4 3 310.00-131207.09-130774.53 -0.02 -0.04 T
0.00
# exchange 6 REMD group 2
2 4 310.50-130974.71-131132.11 -0.01 0.01 F
0.67
3 3 310.50-131192.34-131008.89 -0.01 0.02 T
0.67
4 2 310.50-131008.90-131192.36 -0.01 -0.02 T
0.67
1 1 310.50-131132.10-130974.70 0.02 0.02 F
0.00
# exchange 8 REMD group 2
1 4 310.50-130930.08-131185.88 0.03 0.02 F
1.00
2 3 310.50-131103.90-130812.07 -0.01 0.02 T
1.00
3 2 310.50-130812.07-131103.92 -0.01 0.02 T
1.00
4 1 310.50-131185.93-130930.12 -0.01 -0.04 F
0.00
# exchange 8 REMD group 1
2 2 310.00-130849.89-131080.91 -0.02 0.01 T
1.00
3 1 310.00-131080.93-130849.90 -0.02 0.01 T
0.50
4 4 310.00-131523.30-130969.49 -0.02 -0.03 T
1.00
1 3 310.00-130969.46-131523.28 0.02 0.01 T
0.00
# exchange 10 REMD group 1
1 2 310.00-131420.47-130978.66 0.02 0.01 T
1.20
2 1 310.00-130978.68-131420.48 -0.02 0.01 T
0.80
3 4 310.00-131322.88-130948.52 -0.02 0.01 T
1.20
4 3 310.00-130948.53-131322.88 -0.02 -0.03 T
0.00
# exchange 10 REMD group 2
2 4 310.50-130955.80-130979.16 -0.01 0.02 F
0.80
3 3 310.50-131441.56-130829.72 -0.01 0.02 T
0.80
4 2 310.50-130829.73-131441.56 -0.01 -0.04 T
0.80
1 1 310.50-130979.15-130955.78 0.03 0.02 F
0.00

_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Feb 27 2019 - 18:00:03 PST
Custom Search