AMBER Archive (2003)

Subject: Re: AMBER: Can use mm_pbsa_statistics.pl to calculate each snapshot's Delta Delta G, then give the SD

From: Holger Gohlke (gohlke_at_bioinformatik.uni-frankfurt.de)
Date: Fri Nov 07 2003 - 02:23:29 CST

Zheng Suxin schrieb:
>
> Dear Holger Gohlke,
> Thank for your reply. Followed is my calculation example
>
> compound1
> # DELTA
> # -----------------------
> # MEAN STD
> # =======================
> ELE -219.32 5.45
> VDW -53.23 2.57
> INT 0.00 0.00
> GAS -272.55 6.14
> PBSUR -3.12 0.19
> PBCAL 256.63 6.68
> PBSOL 253.51 6.60
> PBELE 37.31 4.67
> PBTOT -19.05 4.90
>
> compound2
> # DELTA
> # -----------------------
> # MEAN STD
> # =======================
> ELE -210.49 5.40
> VDW -52.82 2.35
> INT 0.00 0.00
> GAS -263.31 5.94
> PBSUR -3.37 0.19
> PBCAL 248.72 6.63
> PBSOL 245.35 6.58
> PBELE 38.23 5.38
> PBTOT -17.96 5.27
>
> then the result should be :delta deltaG= -1.09; SD= 7.20 (SD^2=SD1^2+SD2^2)
>
> Can we ues mm_pbsa_statistics.pl to calculte two sets data, compound1 and compound2, that is six files such as com1_com.all.out, com1_rec.all.out, com1_lig.all.out, com2_com.all.out, com2_rec.all.out, com2_lig.all.out in the statics.in file, to calculate the delta delta G of each snapshots, and give out the mean delta delta G and the STD. This STD I think would be different from the STD estimated as above approach.

No, this doesn't work currently. The only thing you can do is to use one
xxx.all.out file as input (for "stability" calc.) or three (for absolute
binding free energies). In the latter case, you need to distinguish:
1. If com, rec, lig come from ONE trajectory, set the "calc delta"
parameter to 1. This calculates the binding free energy as <com(i) -
rec(i) - lig(i)>, and SD accordingly. 2. If com, rec, lig come from
three separate trajectories, set "calc delta" to 2. This calculates the
binding free energy as <com(i)> - <rec(i)> - <lig(i)>. In this case, the
SD is given as sd_total = sqrt(sd_com^2 + sd_rec^2 + sd_lig^2).

In case of your DDG calculations, you need to proceed as you did above
with respect to the SD.

Best regards

Holger

>
> Best Regards
>
> Zheng Suxin
>
> ŌŚ 03-11-6 8:24:00 ÄśŠ“µĄ£ŗ
> >Zheng Suxin schrieb:
> >>
> >> Dear amber,
> >> When I use MM_PBSA to calculate the delta G, the program would give a SD(standard deviatia) of the delta G.
> >> If I want to get two run's delta delta G, how to estimate the SD.
> >
> >If you want to combine variances (s_u, s_v) of individual parameters (u,
> >v) to get the variance of the result (s_x) and if the fluctuations in
> >the observations of u and v are uncorrelated, the following holds:
> >
> >For x = f(u,v), s_x^2 = s_u^2 * (dx/du)^2 + s_v^2 *(dx/dv)^2.
> >
> >In the case of DDG = DG2 - DG1, you would get: s_DDG^2 = s_DG2^2 +
> >s_DG1^2 for the variance of DDG.
> >
> >(See e.g. P.R. Bevington, Data Reduction and Error Analysis for the
> >Physical Sciences.)
> >
> >> Can MM_PBSA calculates it automately?
> >
> >No.
> >
> >Best regards
> >
> >Holger
> >
> >> Thanks
> >> sxzheng
> >>
> >> -----------------------------------------------------------------------
> >> The AMBER Mail Reflector
> >> To post, send mail to amber_at_scripps.edu
> >> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
> >
> >--
> >++++++++++++++++++++++++++++++++++++++++++++++++++
> >Dr. Holger Gohlke
> >
> >J.W. Goethe-Universität
> >Fachbereich Biologie und Informatik
> >Institut für Mikrobiologie
> >Marie-Curie-Str. 9
> >60439 Frankfurt/Main
> >Germany
> >
> >Tel.: (+49) 69-798-29503; Fax: (+49) 69-798-29826
> >Email: gohlke_at_bioinformatik.uni-frankfurt.de
> >URL: http://www.rz.uni-frankfurt.de/~hgohlke
> >++++++++++++++++++++++++++++++++++++++++++++++++++
> >
> >-----------------------------------------------------------------------
> >The AMBER Mail Reflector
> >To post, send mail to amber_at_scripps.edu
> >To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
> >
> >.
>
> ÖĀ
> Ąń£”
>
> Zheng Suxin
> sxzheng_at_mail.shcnc.ac.cn
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu 

-- 
++++++++++++++++++++++++++++++++++++++++++++++++++
Dr. Holger Gohlke

J.W. Goethe-Universität
Fachbereich Biologie und Informatik
Institut für Mikrobiologie	
Marie-Curie-Str. 9              
60439 Frankfurt/Main            
Germany                         

Tel.:  (+49) 69-798-29503; Fax: (+49) 69-798-29826
Email: gohlke_at_bioinformatik.uni-frankfurt.de
URL:   http://www.rz.uni-frankfurt.de/~hgohlke
++++++++++++++++++++++++++++++++++++++++++++++++++

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu