Hi Senal,
First of all I am no expert in running TI. I wanted to apply it on our medicinal chemistry project, but in the end I gave up. Partially because from some sources it seemed that amber 18 has some problems with soft core potentials and our cluster still did not upgraded to amber 20.
I used to neutralize system in the same way as standard MD. Its always better to have system with small charge (-1 to 1) than not neutralized system that can often be more than 10.
Convergence is very tricky. I was using tools like alchemical_analysis
https://github.com/MobleyLab/alchemical-analysis, pymbar and alchemlyb. In the end the most straightforward (I dont know if most correct) way was to use hamiltonian replica exchange in Amber. It directly provides info about window overlap by exchange frequency and also free energy difference between windows. 10 ns is not so short simulation, so it should be enough, but it depends on your system.
Juraj
On 29. 10. 20 16:16, Delwakkada Liyanage, Senal Dinuka wrote:
> Dear sir,
>
> Thank you very much for your kind reply. In practice, we neutralize a system before any MD simulation. But in absolute free energy calculations, I was thinking we should not neutralize the system because we are anyways decharging the system.
> Also, if I did a 12-point lambda simulation for decharging, do we check the convergence per lambda trajectory or for the combination of lambda trajectories?
> Please let me know your thoughts about these.
> Since I'm only decharging a ligand, I'm sampling a lambda point for 10 ns. What would you suggest should I investigate considering the convergence and correctness of the simulation?
>
> Thank you very much for your time.
>
> Senal
>
> ________________________________
> From: Juraj Dobias <juraj.dobias.uochb.cas.cz>
> Sent: Thursday, October 29, 2020 4:04 AM
> To: amber.ambermd.org <amber.ambermd.org>
> Subject: Re: [AMBER] Amber TI - crgmask
>
> Hi Senal,
>
> Amber will assign atoms that match crgmask zero partial charge, its so
> simple. So in decharging step you basicly have two copies of same ligand
> in your input pdb file and you have to set timask1 to one of them and
> timask2 to second one. Then you have to set crgmask to match part of
> second ligand (the one that you specified by timask2) that you want to
> decharge.
>
> I would recomend you to complete this tutorial
> https://secure-web.cisco.com/18BzDIb__0x_9m-4MOms63gcQmw1O5SbK9D72Slr80DQlw76rLtCfkdU7p_Cv47fRoanpKeCZiU3lRUcWm3SFnQo_we7Ud558hVafjQDSWplpyp-cp4zsvM_0jlIV9sWG9Kb9UU4SRd4pWhv3sYCrRFPlXuMcHQBA_aJPfB7Riv9RN02pD1nTgmxpt7HCknLTPos0qA_Jiry6vaPp5hKOVqTREC6TRpw4rbyeBjYaVqJjDmPk7slhckCA9Bn_IJ7Eqroad-zld_c8MEyE56w39bL33hE4NlWYJ7MKBNTAMhU_M_TjylnLWDDlJDeGFYyV/https%3A%2F%2Fambermd.org%2Ftutorials%2Fadvanced%2Ftutorial9%2Findex.html
>
> And be sure you understand timasks scmasks and crgmask.
>
> Juraj
>
> On 28. 10. 20 17:02, Delwakkada Liyanage, Senal Dinuka wrote:
>> Dear Amber users,
>>
>> I have a doubt about using crgmask flag in thermodynamic integration. In a decharging step, if I use crgmask for my ligand, does that mean my ligand charges are already zero?
>> Also, if my timask1=':ligand_resname', what should be the timask2?
>>
>> Thank you
>>
>> Senal
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://secure-web.cisco.com/1XKhBB4uikVi-i1ve93gWvnpxvFgvhTvwV3xMH4UWUSugOGdryKZM07KNLkvKPzBdpNqz7k-Wu6MxmDzGdMnaB2IWbw0KnkLIrW13BEnrzgbzGbF0IO83NO39WA56WjKyDwpI72Dl7rZOCjNJsm3w1QdpLfXbSFksZG-Uvve3l13Vhjug_lupWDPecedwBDyHtTFiAiVnyuT8u18jO38sQUggTPLvgtV-dUsElExPhwb4tZsGCnGGgHUfnW6OByu_xAwAz-q9CYUTliERTf48vdK8nwUDt3gxHoqdISRwKVPb-J0C8Xk2DfKZ0chLVSLp/http%3A%2F%2Flists.ambermd.org%2Fmailman%2Flistinfo%2Famber
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://secure-web.cisco.com/1XKhBB4uikVi-i1ve93gWvnpxvFgvhTvwV3xMH4UWUSugOGdryKZM07KNLkvKPzBdpNqz7k-Wu6MxmDzGdMnaB2IWbw0KnkLIrW13BEnrzgbzGbF0IO83NO39WA56WjKyDwpI72Dl7rZOCjNJsm3w1QdpLfXbSFksZG-Uvve3l13Vhjug_lupWDPecedwBDyHtTFiAiVnyuT8u18jO38sQUggTPLvgtV-dUsElExPhwb4tZsGCnGGgHUfnW6OByu_xAwAz-q9CYUTliERTf48vdK8nwUDt3gxHoqdISRwKVPb-J0C8Xk2DfKZ0chLVSLp/http%3A%2F%2Flists.ambermd.org%2Fmailman%2Flistinfo%2Famber
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Oct 30 2020 - 04:00:02 PDT
