[claireloison]

Dear all,

I try to learn to use Genesis using the very nice tutorials.

I wish to find some help about the tutorial 3.2, containing a free-energy
calculation for a conformational change of ala-tripeptide.
* Did someone succeeded to run the tutorial 3.2 until the end ? With which version of genesis ?
* Are there some more explanations on the free energy part ?
* Is there some incompatibility with genesis version1.6.0 or genesis1.5.1 and this tutorial ? Which version should I use then ?

Thanks for any help !
Claire Loison

PS : more details on the problems :

I am using genesis1.6 version, compiled with intel2019.4 and associated mpi and mkl.
I get into some errors when using the script run.sh
in the directory 10_analysis/
I tested to use genesis1.5.1 for the analysis only and got the same error.

The free energy is calculated using REUS along a path with 1 CV optimized by r_path
The CV contains 2 constraints on dihedrals. There are 16 walkers.
My feeling is that all the steps until the 9 were ok.

I have difficulties with the analysis of the REUS.
Unfortunately, the explanation on the tutorial are not yet published,
so that debugging is difficult for me.

The first error was using remd_convert. It seems that some necessary options were not in the inp file.
I have added these three lines to the inp file, and changed the one on crdout_period :

nsteps = 100000 # nsteps of Dynamics in reus
num_replicas = 16 # number of replicas
exchange_period = 1000 # exchange periode in reus
crdout_period = 100 # input dcdfile MD period

After that I had the feeling the reordering did his job, and the trj_analysis also.

But then, I had some difficulty with the mbar_analysis. It ended with an error on target temperature.
I tried to add the line to mbar_analysis.inp (or to its template)
target_temperature = 300.0

But I get some bug, and the end of output looks like :
===========================
Read_Ctrl_Option> Parameters of Restraints

# of functions : 2
function : 1
constant = 0.001523 0.001523 0.00152…0.001523 0.001523 0.001523
reference = -0.63245E+02 -0.70423E+02… -0.66063E+02 -0.62695E+02
is_periodic = yes
box_size = 360.000

function : 2
constant = 0.001523 0.001523 0.00152…0.001523 0.001523 0.001523
reference = -0.41057E+02 -0.29662E+02…03 0.12986E+03 0.14290E+03
is_periodic = yes
box_size = 360.000

[STEP2] Set Relevant Variables and Structures

Setup_Option> Unit for fene and pmf is set to dimension less

[STEP3] Analysis trajectory files

Build_Data_K_From_Cv>
number of time steps : 1000

read cv file: 1.tor
read cv file: 2.tor
read cv file: 3.tor
read cv file: 4.tor
read cv file: 5.tor
read cv file: 6.tor
read cv file: 7.tor
read cv file: 8.tor
read cv file: 9.tor
read cv file: 10.tor
read cv file: 11.tor
read cv file: 12.tor
read cv file: 13.tor
read cv file: 14.tor
read cv file: 15.tor
read cv file: 16.tor
..done

Build_U_Kl_From_CV>
..done

Solve_Mbar>

[block 1] 1th iteration delta = NaN tolerance = 0.100000000E-06
[block 1] 2th iteration delta = NaN tolerance = 0.100000000E-06
[block 1] 3th iteration delta = NaN tolerance = 0.100000000E-06
[block 1] 4th iteration delta = NaN tolerance = 0.100000000E-06

===========================

[ckobayashi]

Dear claireloison,

From your message, I think you tried the tutorial for previous versions (1.0-1.3). Please use the newer version of tutorial for recent versions (1.4 or later) of GENESIS.

https://www.r-ccs.riken.jp/labs/cbrt/tutorials2019/

You can also go there directly from the menu (“Tutorials&Sample”) at the top of the site.

The path sampling tutorial is here;

https://www.r-ccs.riken.jp/labs/cbrt/tutorials2019/genesis-tutorial-14-1/

Sincerely yours,

Kobayashi