Transcript Ligand preparation in practice

Ligand preparation in
practice
Balázs Jójárt
Directory & set path
 in SS2015 make the
 2.ligprep directory
 21.test
 copy from the cluster the output (C2H4O2.sdf file) into
21.test directory
 launch Mestro
 use Project/Change Directory… & set the
SS2015/2.ligprep/21.test as the destination directory
Import structure into Maestro
 Project/Import Structures… OR Crtl+i
 select sdf file: C2H4O2.sdf & Open
 On the left hand site you can see the „Entry list”
 All structures are selected: yellow foreground but not displayed
 Clicking on the second column  display structure
 Shift/Ctrl+clicking on second boxes  more structures are displayed
 Clicking into 3rd column: enter the ID of the molecule
 please enter from 1 .. 10
 if we close Maestro all IDs will lost  we save into a project
 Project/Save as  test.ligprep.prj
Ligprep C2H4O2 I.
 what do you think, how many other structures can we generate with
ligprep
 don’t forget: we can set up different ionization states and configuration
 before we start
 Title: change C2H4O2 to C2H4O2.init
 Tasks/Application
 Tasks: Tasks/Application View
 Application View
 LigPrep…
Ligprep C2H4O2 II.
change to: Project Table
(Selected Entries don’t forget
to select all structures!!!)
EPIK: predict most probable
protonation state @ given pH
no DESALT & no TAUTOMERS
Generate all combinations:
32  25  5 chirality center
for every ring 1 low energy
conformation
sdf & C2H4O2.ligprep
Ligprep C2H4O2 III
Ligprep C2H4O2 IV.
monitor the job
Applications/Monitor Jobs…
take a closer look into the log file!
C2H4O2.ligprep directory
contains all input and output files
check structures with ID 7
superposition: right hand side: super 
Tools/Superpostion
Selected Entries
ASL/Select…  Click on the atoms in the ring
Ligprep PET I.
what about your petmolecules?
there are more than 100 possible structures
how include ID? (by hand ;))
10 molecules  ~ 30 seconds
for PET all calculations will performed on the
cluster
Ligprep PET II.
 open putty & login cluster
 cd SS2015
 mkdir 2.ligprep && cd 2.ligprep
 mkdir 22.pet && cd 22.pet
 launch mc
 copy into 22.pet your sdf output file from 1.molgen/12.pet
 copy convert.sh & insert.ID.sh into 22.pet (/tmp/SS2015/scripts)
 cd /tmp
 bash convert.sh &> convert.out &
 structconvert: converter between various formats: sdf  smi
 ? smi: easier to include structure ID
 bash insert.ID.sh &> insert.ID.out &
 in 2nd column ID numbers
Ligprep PET III.
please check the numbers
010_C10H16NO-NH3+.id.smi  2618
012_C10H16NO2-NH3+.id.smi  18708
013_C10H16NO3-NH3+. id.smi  100060
020_C08H12NO3-NH3+. id.smi  4028
021_C08H12NO1-NH3+. id.smi  175
022_C08H12NO2-NH3+. id.smi  962
030_C09H14NO3-NH3+. id.smi  21424
031_C09H14NO1-NH3+. id.smi  709
032_C09H14NO2-NH3+. id.smi  4487
PET==stoichometry!!!!!!
Ligprep PET IV.
 original sdf and converted smi files are not yet necessary  DELETE
 DON’T DELETE PET.id.smi !!!!!
 on your laptop there is an input file: C2H4O2.ligprep.inp
 copy this file into 22.pet directory
 rename it: mv C2H4O2.ligprep.inp PET.ligprep.inp
 instead: C2H4O2.ligprep.maegz  PET.id.smi
 ls -1 PET.id.smi >> PET.ligprep.inp
 F4, go to beginning of tle last line
 F3 & END & F3
 in the first row: delete the *.maegz & F6
 in second row delete *.sdf & F5 and CHANGE SMI to SDF at the end!!!!
 F2: save & F10: quit
Ligprep PET V.
 we have to submit the job to the nodes
 copy ligprep.qsub into SS2015/2.ligprep/22.pet
 change F4
 NP as follows
 010_C10H16NO-NH3+.id.smi  2618  1
 012_C10H16NO2-NH3+.id.smi  18708  8
 013_C10H16NO3-NH3+. id.smi  100060
 020_C08H12NO3-NH3+. id.smi  4028  2
 021_C08H12NO1-NH3+. id.smi  175 1
 022_C08H12NO2-NH3+. id.smi  962  1
 030_C09H14NO3-NH3+. id.smi  21424  10
 031_C09H14NO1-NH3+. id.smi  709  1
 032_C09H14NO2-NH3+. id.smi  4487  2
Ligprep PET VI.
F2 & F10
qsub ligprep.qsub
qstat –u USERNAME
Into report from Ligprep
 Materials & Methods
 steps performed in order to obtain structures
 Results & Discussion
 number of original structures (from PET.ID.smi)
 ? structures was generated byEPIK?
 structures was generated by stereoizer
 was premin for all structures OK?
 finally how many structures were generated from the original smi
file
Protein Preparation Wizard I.
 proteins
 contain several tens or hundreds of amino acids
 don’t have Hs
 protonation states are very important properties of amino acids
 protonation states are pH dependent
 @ pH = 7
 carboxylate groups (C-terminal, Asp, Glu)
 protonated N (N-terminal, Lys, Arg)
 protonated OH (Tyr)
 interesting case is His 6.5 is very close to 7
http://physiologyonline.physiology.org/content/nips/22/1/30/F2.large.jpg
Protein Preparation Wizard II.
 we don’t have a WAND ;)
 we have to use a program to correct protein structures
 Protein Preparation Wizard
 add missing Hs
 add missing heavy atoms, side chains (REMARK 465 and REMARK
470)
 predict ionization state of the ligand
 protonation states of amino acids in the receptor
 final optimize the structure
Protein Preparation Wizard III.
make a directory in SS2015  3.protprep
launch Maestro
Project/Change Directory… & choose 3.protprep
Get PDB
3sn6 & R chain
delet unnecessary part
and fit the structure into the center of the workspace
use Prep Wiz button
and wait a little bit ;)
Protein Preparation Wizard IV.
change prepwizard to 3sn6
Align to: 0
Assign bond orders: 1
Add hydrogens: 1
Create zero-order … : 0
Create disulfide bonds: 1
Convert selevno … : 0
Fill missing side chains …: 1
Fill missing loops …: 0
Cap termini: 0
Delete water: 0
Protein Preparation Wizard V.
 we can monitor the job
 take a closer look on
 ligand structures
 surrounding amino acids
 Review and Modify
 Generate States
 Epik calculation for ligand @ pH 7 ± 3
 display only ligand in the workspace
 select between various states
 state penalty: Orig  3.13 ; S2  0.66 ; S3  3.91 and S4  4.02
kcal/mol
 S2 is the most likely structure in solution
Protein Preparation Wizard VI.
 Refine tab
Protein Preparation Wizard VII.
 Refine tab
 visualize ligand + protein amino acids in 3 Å
 check the protonation state of the Asp113!!!
Protein Preparation Wizard VIII.
 IF:
 Running PropKa to get pKas
 deprotonated Asp113
 THEN
don’t forget to save a
project file!!!
3sn6.prep.prj
Into report from ProtPrepWizard
Materials & Methods
steps performed in order to obtain structures
Results & Discussion
protonation state of the ligand
protonation state of surrounding amino acids
a nice picture about interactions