Emilio,
I continue my previous email:
See again
http://q4md-forcefieldtools.org/Tutorial/leap-mol2.php
using the simple leap script I provided below:
RBFLV = sequence {LNO FLO}
select RBFLV.2
relax RBFLV
deselect RBFLV.2
savemol2 RBFLV RBFLV.mol2 1
# obviously atom types & FF parameters have to be defined/loaded
before relaxing...
-> this solves the problem...
The script: http://q4md-forcefieldtools.org/REDDB/projects/F-90/script1.ff
can be easily adapted to implement a 'relax' job for all the FLO units...
RBFLV = seq {LNO FLO}
RBFLVH = seq {LNO FLR}
FMN1 = seq {P1M LNK FLO}
FMN1H = seq {P1M LNK FLR}
FMN2 = seq {P2M LNK FLO}
FMN2H = seq {P2M LNK FLR}
select RBFLV.2
relax RBFLV
deselect RBFLV.2
# This is strange: no problem with FLR ;-)
select FMN1.3
relax FMN1
deselect FMN1.3
select FMN2.3
relax FMN2
deselect FMN2.3
etc... for all the other units;
I am going to update script1.ff in REDDB & report that correction at:
http://q4md-forcefieldtools.org/REDDB/error.php
You will be able to use it for FAD and other cofactors as well...
I hope this helps...
regards, Francois
----- Forwarded message from fyd.q4md-forcefieldtools.org -----
Date: Sat, 01 Nov 2014 21:50:17 +0100
From: FyD <fyd.q4md-forcefieldtools.org>
Reply-To: AMBER Mailing List <amber.ambermd.org>
Subject: Re: [AMBER] [q4md-fft] FAD wrong geometry in F-90 project
To: q4md-fft.q4md-forcefieldtools.org, Emilio Lence
<guitarro.de.meixente.gmail.com>
Cc: AMBER Mailing List <amber.ambermd.org>
Dear Emilio,
First you are right there is a problem; my feeling is that this is
just 'another' LEaP bug; this is why fragment fusion can now directly
be carried out by PyRED (R.E.D. Server Dev.
http://q4md-forcefieldtools.org/REDServer-Development/).
Then, this problem of wrong Cartesian coordinates does not matter in
the context of a force field library (_IF_ the topology is correct);
because usually the Cartesian coordinates from the FF lib. are
replaced by these of the PDB file (experimental data);
see See
http://q4md-forcefieldtools.org/Tutorial/leap-mol2.php
also after geometry optimization the problem should be solved; my
feeling is that you can carry on your procedure...
Going back to the problem you reported:
You can see that the Cart. coordinates are correct for the two
isolated fragments before fragment linking and the problem appends
after linking the head and tail of these 2 fragments using the
'sequence' command...
This is known that LEaP has bugs (to my knowledge; they remains
uncorrected) when linking the head/tail of two fragments in some
cases; here this is quite surprising because the connection between
the head/tail seems to be correctly done, but the geometry is modified
at another place of the molecule...
A workaround could be to invert head/tail into tail/head; although
once again the Cart. coordinates in the FF lib. do not matter if the
topology is correct.
I still work on this problem to figure out what's going on... I
simplified the problem and provided below the leap script and the 2
mol2 files (two fragments). May be someone in the list can have some
ideas and help...
regards, Francois
--- the leap script---
logFile leap.log
LNO = loadmol2 ./lno.mol2
FLO = loadmol2 ./flo.mol2
verbosity 2
addAtomTypes {
{ "H" "H" "sp3" }
{ "HO" "H" "sp3" }
{ "HS" "H" "sp3" }
{ "H1" "H" "sp3" }
{ "H2" "H" "sp3" }
{ "H3" "H" "sp3" }
{ "H4" "H" "sp3" }
{ "H5" "H" "sp3" }
{ "HW" "H" "sp3" }
{ "HC" "H" "sp3" }
{ "HA" "H" "sp3" }
{ "HP" "H" "sp3" }
{ "OH" "O" "sp3" }
{ "OS" "O" "sp3" }
{ "O" "O" "sp2" }
{ "O2" "O" "sp2" }
{ "OW" "O" "sp3" }
{ "CT" "C" "sp3" }
{ "CH" "C" "sp3" }
{ "C2" "C" "sp3" }
{ "C3" "C" "sp3" }
{ "C" "C" "sp2" }
{ "C*" "C" "sp2" }
{ "CA" "C" "sp2" }
{ "CB" "C" "sp2" }
{ "CC" "C" "sp2" }
{ "CN" "C" "sp2" }
{ "CM" "C" "sp2" }
{ "CK" "C" "sp2" }
{ "CQ" "C" "sp2" }
{ "CD" "C" "sp2" }
{ "CE" "C" "sp2" }
{ "CF" "C" "sp2" }
{ "CP" "C" "sp2" }
{ "CI" "C" "sp2" }
{ "CJ" "C" "sp2" }
{ "CW" "C" "sp2" }
{ "CV" "C" "sp2" }
{ "CR" "C" "sp2" }
{ "CA" "C" "sp2" }
{ "CY" "C" "sp2" }
{ "C0" "C" "sp2" }
{ "MG" "Mg" "sp3" }
{ "N" "N" "sp2" }
{ "NA" "N" "sp2" }
{ "N2" "N" "sp2" }
{ "N*" "N" "sp2" }
{ "NP" "N" "sp2" }
{ "NQ" "N" "sp2" }
{ "NB" "N" "sp2" }
{ "NC" "N" "sp2" }
{ "NT" "N" "sp3" }
{ "N3" "N" "sp3" }
{ "S" "S" "sp3" }
{ "SH" "S" "sp3" }
{ "P" "P" "sp3" }
}
parm99 = loadamberparams parm99.dat
frcmod99SB = loadamberparams frcmod.ff99SB
#
http://q4md-forcefieldtools.org/REDDB/projects/F-90/script3.ff
FRCMOD = loadamberparams script3.ff
set LNO name "LNO"
set LNO head LNO.1.O5
set LNO tail LNO.1.C1
set LNO.1 connect0 LNO.1.O5
set LNO.1 connect1 LNO.1.C1
set LNO.1 restype undefined
set LNO.1 name "LNO"
# set FF atom types
set LNO.1.C1 type CT
set LNO.1.H11 type H1
set LNO.1.H12 type H1
set LNO.1.C2 type CT
set LNO.1.H2 type H1
set LNO.1.O2 type OH
set LNO.1.HO2 type HO
set LNO.1.C3 type CT
set LNO.1.H3 type H1
set LNO.1.O3 type OH
set LNO.1.HO3 type HO
set LNO.1.C4 type CT
set LNO.1.H4 type H1
set LNO.1.O4 type OH
set LNO.1.HO4 type HO
set LNO.1.C5 type CT
set LNO.1.H51 type H1
set LNO.1.H52 type H1
set LNO.1.O5 type OH
set LNO.1.HO5 type HO
set FLO name "FLO"
set FLO head FLO.1.N10
set FLO tail FLO.1.N1
set FLO.1 connect0 FLO.1.N10
set FLO.1 connect1 FLO.1.N1
set FLO.1 restype undefined
set FLO.1 name "FLO"
# set FF atom types
set FLO.1.N1 type NC
set FLO.1.C2 type C
set FLO.1.O2 type O
set FLO.1.N3 type NA
set FLO.1.H3 type H
set FLO.1.C4 type C
set FLO.1.O4 type O
set FLO.1.N10 type "N*"
set FLO.1.C10A type CB
set FLO.1.C4A type CB
set FLO.1.N5 type NC
set FLO.1.C5A type CB
set FLO.1.C9A type CB
set FLO.1.C9 type CA
set FLO.1.H9 type HA
set FLO.1.C8 type CA
set FLO.1.C7 type CA
set FLO.1.C6 type CA
set FLO.1.H6 type HA
set FLO.1.C7M type CT
set FLO.1.H7M1 type HC
set FLO.1.H7M2 type HC
set FLO.1.H7M3 type HC
set FLO.1.C8M type CT
set FLO.1.H8M1 type HC
set FLO.1.H8M2 type HC
set FLO.1.H8M3 type HC
RBFLV = sequence {LNO FLO}
savemol2 RBFLV RBFLV.mol2 1
# the geometry in lno.mol2 & flo.mol2 is correct
# the geometry is modified in RBFLV.mol2
# the head/tail connection looks correct
# bad distance : 0.76 ang. between atoms C4A & N5
-------------------------------------------------------
the two mol2 files:
.<TRIPOS>MOLECULE
FLO
27 29 1 0 1
SMALL
USER_CHARGES
.<TRIPOS>ATOM
1 N1 2.533395 -1.286901 0.000036 N 1 FLO
-0.5665
2 C2 3.736048 -0.619349 0.000020 C 1 FLO
0.6967
3 O2 4.788429 -1.175183 0.000036 O 1 FLO
-0.5640
4 N3 3.712301 0.777897 -0.000015 N 1 FLO
-0.4297
5 H3 4.606249 1.222732 -0.000025 H 1 FLO
0.3208
6 C4 2.617587 1.592059 -0.000037 C 1 FLO
0.4131
7 O4 2.686375 2.779343 -0.000067 O 1 FLO
-0.5067
8 N10 0.244333 -1.306996 0.000034 N 1 FLO
-0.0419
9 C10A 1.426559 -0.638229 0.000018 C 1 FLO
0.2477
10 C4A 1.329041 0.829082 -0.000019 C 1 FLO
0.4606
11 N5 0.236577 1.467844 -0.000037 N 1 FLO
-0.5248
12 C5A -0.944773 0.770411 -0.000021 C 1 FLO
0.2744
13 C9A -0.961816 -0.622077 0.000015 C 1 FLO
-0.0030
14 C9 -2.199470 -1.272402 0.000030 C 1 FLO
-0.3116
15 H9 -2.246149 -2.342434 0.000057 H 1 FLO
0.1782
16 C8 -3.379999 -0.558008 0.000010 C 1 FLO
0.1153
17 C7 -3.363362 0.855346 -0.000026 C 1 FLO
0.0374
18 C6 -2.146015 1.486519 -0.000040 C 1 FLO
-0.2174
19 H6 -2.074732 2.558453 -0.000067 H 1 FLO
0.1851
20 C7M -4.645567 1.654317 -0.000048 C 1 FLO
-0.2294
21 H7M1 -4.433476 2.716128 -0.000074 H 1 FLO
0.0829
22 H7M2 -5.251623 1.438489 -0.874794 H 1 FLO
0.0829
23 H7M3 -5.251625 1.438534 0.874709 H 1 FLO
0.0829
24 C8M -4.693207 -1.303067 0.000027 C 1 FLO
-0.2117
25 H8M1 -5.285556 -1.051151 0.874030 H 1 FLO
0.0806
26 H8M2 -5.285554 -1.051195 -0.873990 H 1 FLO
0.0806
27 H8M3 -4.536776 -2.374245 0.000055 H 1 FLO
0.0806
.<TRIPOS>BOND
1 1 2 1
2 1 9 1
3 2 3 1
4 2 4 1
5 4 5 1
6 4 6 1
7 6 7 1
8 6 10 1
9 8 9 1
10 8 13 1
11 9 10 1
12 10 11 1
13 11 12 1
14 12 13 1
15 12 18 1
16 13 14 1
17 14 15 1
18 14 16 1
19 16 17 1
20 16 24 1
21 17 18 1
22 17 20 1
23 18 19 1
24 20 21 1
25 20 22 1
26 20 23 1
27 24 25 1
28 24 26 1
29 24 27 1
.<TRIPOS>SUBSTRUCTURE
1 FLO 1 **** 0 **** ****
-----------------------------------------------------------------
.<TRIPOS>MOLECULE
LNO
20 19 1 0 1
SMALL
USER_CHARGES
.<TRIPOS>ATOM
1 C1 -2.520439 0.630546 0.199687 C 1 LNO
-0.0599
2 H11 -2.532568 0.715827 1.281617 H 1 LNO
0.1278
3 H12 -2.466342 1.636330 -0.193799 H 1 LNO
0.1278
4 C2 -1.273437 -0.158446 -0.201527 C 1 LNO
0.2676
5 H2 -1.222264 -0.233246 -1.289744 H 1 LNO
0.0329
6 O2 -1.404734 -1.445262 0.357788 O 1 LNO
-0.6359
7 HO2 -0.975801 -2.073417 -0.205478 H 1 LNO
0.3898
8 C3 0.024691 0.498335 0.293316 C 1 LNO
0.0054
9 H3 0.021452 0.443797 1.377208 H 1 LNO
0.1262
10 O3 0.050908 1.873155 -0.010127 O 1 LNO
-0.6751
11 HO3 0.054518 1.995838 -0.950945 H 1 LNO
0.4289
12 C4 1.301779 -0.162854 -0.257700 C 1 LNO
0.1783
13 H4 1.349408 0.042496 -1.322420 H 1 LNO
0.1191
14 O4 1.277186 -1.563722 -0.158210 O 1 LNO
-0.6201
15 HO4 1.080505 -1.812016 0.735982 H 1 LNO
0.4122
16 C5 2.551048 0.412668 0.392132 C 1 LNO
0.0157
17 H51 2.603965 0.060669 1.423047 H 1 LNO
0.0752
18 H52 2.485623 1.494853 0.411653 H 1 LNO
0.0752
19 O5 3.663788 -0.014852 -0.341979 O 1 LNO
-0.6269
20 HO5 4.455044 0.311871 0.061845 H 1 LNO
0.4227
.<TRIPOS>BOND
1 1 2 1
2 1 3 1
3 1 4 1
4 4 5 1
5 4 6 1
6 4 8 1
7 6 7 1
8 8 9 1
9 8 10 1
10 8 12 1
11 10 11 1
12 12 13 1
13 12 14 1
14 12 16 1
15 14 15 1
16 16 17 1
17 16 18 1
18 16 19 1
19 19 20 1
.<TRIPOS>SUBSTRUCTURE
1 LNO 1 **** 0 **** ****
-----------------------------------------------------------------
> I recently start to work with a protein that has FAD and NADP as cofactors.
> After a search I found your parameters, so I setup my system nicely, with a
> couple of minor hiccups. But when I looked into the mol2 file for FAD, I
> encountered that the geometry is wrong. I checked the topoloy and
> coordinate files.and has the same error, and the same happened in the
> dFAD.mol2, FADP1.mol2, FADP2.mol2 files. Those are the ones I checked.
> The error consist in a very short bond distance of 0.75579 between atoms
> C4A and N5 (unit FLO). That value clearly is wrong.
> What I want to know is if that mistake in the bond distance is something
> that happened when you uploaded the parameters or make the scripts, etc.;
> so then is not going to make anything funny in my MD simulations. Or is
> this geometry used in the calculations of the charges, and then I can
> expect something wrong happen.
> I used the coordinates of the X-ray structure, so my initial geometry for
> the FAD has not that problem, hoping that anything will run smoothly.
----- End forwarded message -----
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Nov 01 2014 - 23:00:01 PDT
