changing thrombin model

calculating model2 with sodium

calculating model2 without sodium

calculating model2 with sodium

2. Building a receptor model

For the first thrombin model 1c5l (human thrombin, apo) was selected, but no correlation between predicted DG values for docking solutions and the corresponding experimental values could be found. It's known that rigid docking into apo structures leads often to wrong results (- Leslie A. Kuhn at Keystone Symposia: only 9 of 42 thrombin ligands could be docked into the apo structure with a rigid protein model.)

The minimization of crystal structure 1dwd, 3 A resolution, crystallized together with NAPAP (Banner Hadvary, JBC 256, 1991; Ki = 6.6 nm) as an equivelent to bovine thrombin 1ets (Bode Brandstetter; also with NAPAP) does not work for unkown reasons.

The receptor model was changed to the human thrombin non-apo structure 1c4u - A89 (R.KRISHNAN,I.MOCHALKIN,R.K.ARNI,A.TULINSKY; 2.1 A; no electron density map is available). Ki=4.4 fM (pKi -10.89, JCICS 2004,44,1872).

copy sodium from 1c4u.pdbf
TYS 567 changed to TYR 567 by removing the sulfate group. - thrombin-model-1c4u-temp.pdb

xleap.sh -pdb thrombin-model-1c4u -n 000 -prot thrombin-model-1c4u-temp.pdb -lig model1.ligand.pdb -m /home/henricsn/combine2go/parameter/protmin.thrombin-model.template -w -noac -nopc -notemp -bres |tee xleap2.log

Minimization together with ligand A89:
xleap.sh -pdb thrombin-model2-1c4u -n A89 -prot thrombin-model-1c4u-temp.pdb -lig 1c4u.ligand.pdb -m /home/henricsn/combine2go/parameter/protmin.medium_constrain.template -w -notemp -bres -sdf 1c4u.sdf |tee xleap3.log
(Amidino group and ring system not planar!)

removing ligand and water receptor model:
thrombin-model2-1c4u.minbres.pdb
thrombin-model2-1c4u.min.pdb

3. SD file

checking bonds and hydrogens in ligands /home/henricsn/combine2go/sdf/all_120406_3D.sdf

4. Preparation of ligands

31 ligands:
1a4w.ligand.pdb 1bmn.ligand.pdb 1d9i.ligand.pdb 1ghx.ligand.pdb 1o2g.ligand.pdb 7kme.ligand.pdb
1ae8.ligand.pdb 1c4u.ligand.pdb 1dwc.ligand.pdb 1ghy.ligand.pdb 1o5g.ligand.pdb
1afe.ligand.pdb 1c4v.ligand.pdb 1dwd.ligand.pdb 1gj4.ligand.pdb 1qbv.ligand.pdb
1bcu.ligand.pdb 1c4y.ligand.pdb 1fpc.ligand.pdb 1gj5.ligand.pdb 1ta2.ligand.pdb
1bhx.ligand.pdb 1c5n.ligand.pdb 1ghv.ligand.pdb 1k21.ligand.pdb 1ta6.ligand.pdb
1bmm.ligand.pdb 1d6w.ligand.pdb 1ghw.ligand.pdb 1k22.ligand.pdb 1tom.ligand.pdb

5. Minimization

start minimization in /home/henricsn/combine2go/data/thrombin/model2_180406
local_start.sh -sdf /home/henricsn/combine2go/sdf/all_120406_3D.sdf -inrec ./ -inlig /home/henricsn/combine2go/data/thrombin/pdb/ligand/ -model thrombin-model2-1c4u.min.pdb -ligbase .ligand.pdb |tee local.log

problems with following liands:
1ta6 (A69) antechamber fails because of charge

6. Interaction energies

selected ligands:
A04.0.min.pdb A17.0.min.pdb A45.0.min.pdb A68.0.min.pdb A86.0.min.pdb A91.0.min.pdb
A09.0.min.pdb A18.0.min.pdb A46.0.min.pdb A72.0.min.pdb A87.0.min.pdb A92.0.min.pdb
A12.0.min.pdb A19.0.min.pdb A47.0.min.pdb A73.0.min.pdb A88.0.min.pdb A93.0.min.pdb
A12.1.min.pdb A22.0.min.pdb A48.0.min.pdb A82.0.min.pdb A89.0.min.pdb A94.0.min.pdb
A14.0.min.pdb A44.0.min.pdb A63.0.min.pdb A84.0.min.pdb A90.0.min.pdb

mkdir anal
cd anal/

start calculation of interaction energies
calc_inter_model.sh -ki thrombin -sdf /home/henricsn/combine2go/sdf/all_120406_3D.sdf -indir ../ -outdir ./ -x |tee calc.log

files

min.pdb
topology
xyz
parm
prep

problems with following minimized ligands:
ID residue netcharge thrombin PDB_thrombin
21 A04 1 0.76 1o2g OK
13 A09 1 8.5 1ghy Benzimidazole 5-ring end carbon not flat in minimized structure, should be flat acc. To CSD
10 A12 1 63 1ghw_1ghx OK
20 A14 1 11 1gj5 OK
8 A17 1 11 1o5g OK
2 A18 1 85 1gj4 OK
288 A19 1 20.4 1c5n OK
289 A20 1 29 1c5o not available
291 A22 1 45 1ghv OK
312 A44 1 10 1bcu OK
313 A45 1 0.15 1bhx sulphone amide and guanidine nitrogens not flat in minimized
314 A46 1 0.08 1bmm sulphone amide guanidine nitrogens not flat
316 A47 1 0.01 1ets_1dwd not flat sulphone amide in minimized
315 A48 1 0 1bmn sulphone amide guanidine nitrogens not flat
317 A49 1 2.3 1c1u not available
318 A58 2 3.7 1c1v not available
322 A62 1 1.12 1d6w A62 not available
323 A63 1 0 1d9i X-ray not quite flat amide, minimized not flat 5-ring, look up!!
328 A68 1 0 1ta2 OK
329 A69 1 0 1ta6 A69 nto available
332 A72 0 0 1k21 OK
333 A73 0 0 1k22 OK
342 A82 1 4.17 1qbv OK
344 A84 2 40.74 7kme OK
346 A86 0 0 1dwc X-ray not flat carboxylate!! Minimized not flat guanidine, sulphone amide, P3 nitrogen
347 A87 1 0.28 1ae8 OK
348 A88 1 16.22 1afe 5-ring flat in X-ray structure, not in minimized
349 A89 1 0 1c4u OK
350 A90 1 0 1c4v 5-ring flat in X-ray, not in minimized. Amidine nitrogen
351 A91 1 0.01 1c4y 5-ring flat in X-ray, not in minimized. end nitrogen
352 A92 1 0.1 1fpc P2 amide nitrogen in sp2 in minimized, not flat in X-ray but X-ray is probably wrong
353 A93 1 1.2 1a4w OK
354 A94 1 0.01 1tom positive amine nitrogen, giving total charge +2??

golpe


Data file : /home/henricsn/combine2go/data/thrombin/model2_180406/anal/output.dat

Comment : Golpe dat for COMBINE analysis, 04/18/06 14:12:41

Number of variables = 592
Number of experiments = 29

Number of X-variables = 584 , Y-variables = 1
Loading :

1 A04.0
2 A72.0
3 A91.0
4 A93.0
5 A12.0
6 A14.0
7 A90.0
8 A89.0
9 A48.0
10 A46.0
11 A84.0
12 A94.0
13 A68.0
14 A82.0
15 A73.0
16 A92.0
17 A47.0
18 A86.0
19 A45.0
20 A44.0
21 A88.0
22 A87.0
23 A19.0
24 A22.0
25 A12.1
26 A09.0
27 A17.0
28 A63.0
29 A18.0

Active X-variables (SS 1.0E-7) = 234
Active Y-variables (SS 1.0E-7) = 1

Active X-variables after PRETREATMENT = 234
Active Y-variables after PRETREATMENT = 1


Principal Component Analysis (PCA) 29 objects 234 X-var

components XVarExp XAccum
1 43.6226 43.6226
2 18.0077 61.6303
3 14.0356 75.6659
4 6.3683 82.0342
5 2.8544 84.8886
6 2.1647 87.0532
7 2.1106 89.1638
8 2.0639 91.2278
9 1.4685 92.6963
10 1.1286 93.8249



PCA Rank Validation - using 4 random groups

components PRESS Seps R
1 3.3829e+04 2.9585e+03 11.4345
2 6.4738e+04 1.6056e+03 40.3207
3 8.9832e+02 1.0505e+03 0.8552
4 5.9513e+02 6.3948e+02 0.9306
5 5.0163e+02 4.5246e+02 1.1087
6 2.0829e+03 3.6409e+02 5.7209
7 7.0572e+02 2.9786e+02 2.3693
8 8.5772e+02 2.3756e+02 3.6106
9 2.0243e+02 1.8283e+02 1.1072
10 1.6528e+02 1.4436e+02 1.1449


Partial Least Squares (PLS) 29 objects 234 X-var 1 Y-var

Y1 components XVarExp XAccum SDEC r2
0 0.0000 0.0000 2.6630 0.0000
1 38.2511 38.2511 1.9903 0.4414
2 17.7418 55.9929 1.5430 0.6643
3 4.2975 60.2904 1.1578 0.8110
4 16.7251 77.0155 1.0804 0.8354
5 5.9356 82.9511 1.0222 0.8527
6 2.7692 85.7203 0.9302 0.8780
7 0.9862 86.7065 0.7724 0.9159
8 1.2089 87.9154 0.6852 0.9338
9 1.3134 89.2288 0.6249 0.9449
10 0.3669 89.5956 0.5338 0.9598


PLS Model Validation - LOO

Y1 components SDEP SDEV(sdep) q2
0 2.7581 - -0.0727
1 2.2366 - 0.2946
2 2.1227 - 0.3646
3 2.0077 - 0.4316
4 2.0242 - 0.4222
5 2.0355 - 0.4157
6 2.1979 - 0.3188
7 2.7272 - -0.0488
8 2.9576 - -0.2335
9 2.9562 - -0.2323
10 3.1991 - -0.4432


Deleted unselected vars. (D-Optimal)
1 : 234 - 117 Comp.=3 , Tue Apr 18 17:17:30 2006
Active X-variables after VAR. SELECT. = 117


*** FFD Variable Selection Started ***
Max. dimensionality : 3
Validation Mode : LOO
Recalculate weights : yes
Comb./Var. ratio : 2.0
Use groups :
Uncertains : Retain
Fold-over design : no
perc. of dummies : 20


Deleted unselected vars. (F.Factorial)
1 : 117 - 111 Comp.=3 , Tue Apr 18 17:17:50 2006
Active X-variables after VAR. SELECT. = 111



Principal Component Analysis (PCA) 29 objects 111 X-var

components XVarExp XAccum
1 49.9261 49.9261
2 11.2211 61.1473
3 6.7593 67.9066
4 5.4380 73.3446
5 5.0876 78.4322



PCA Rank Validation - using 4 random groups

components PRESS Seps R
1 6.3996e+02 1.1096e+03 0.5767
2 5.6084e+02 5.3487e+02 1.0486
3 4.5819e+02 3.9895e+02 1.1485
4 3.4505e+02 3.1632e+02 1.0908
5 1.1179e+03 2.5178e+02 4.4398
6 2.1396e+02 1.9490e+02 1.0978
7 1.6279e+02 1.5328e+02 1.0620
8 1.5656e+02 1.2364e+02 1.2663
9 1.0848e+02 9.9415e+01 1.0912
10 8.6756e+01 7.9640e+01 1.0894


Partial Least Squares (PLS) 29 objects 111 X-var 1 Y-var

Y1 components XVarExp XAccum SDEC r2
0 0.0000 0.0000 2.6630 0.0000
1 49.4095 49.4095 1.6670 0.6081
2 9.4836 58.8931 1.3091 0.7583
3 6.5987 65.4919 1.1853 0.8019
4 3.7883 69.2802 1.0974 0.8302
5 3.7179 72.9981 1.0125 0.8555


PLS Model Validation - LOO

Y1 components SDEP SDEV(sdep) q2
0 2.7581 - -0.0727
1 1.8509 - 0.5169
2 1.8043 - 0.5409
3 1.9169 - 0.4819
4 2.1653 - 0.3389
5 2.6461 - 0.0126
6 3.1402 - -0.3905
7 3.4286 - -0.6577
8 3.5913 - -0.8187
9 3.6294 - -0.8574
10 3.7669 - -1.0009

calculating model2 without sodium

remove sodium from thrombin-model2-1c4u.min.pdb - thrombin-model2-1c4u_delNA.min.pdb

local_start.sh -sdf /home/henricsn/combine2go/sdf/all_120406_3D.sdf -inrec ./ -inlig /home/henricsn/combine2go/data/thrombin/pdb/ligand/ -model thrombin-model2-1c4u_delNA.min.pdb -ligbase .ligand.pdb |tee local.log


mkdir anal
cd anal/

start calculation of interaction energies
calc_inter_model.sh -ki thrombin -sdf /home/henricsn/combine2go/sdf/all_120406_3D.sdf -indir ../ -outdir ./ -x |tee calc.log


 Privacy Imprint