SLIDE 1
Chemical properties that afgect binding of enzyme-inhibiting drugs to enzymes
Research proposal by Dan Nacu
Chemical properties that a fg ect binding of enzyme-inhibiting drugs - - PowerPoint PPT Presentation
Chemical properties that a fg ect binding of enzyme-inhibiting drugs - - PowerPoint PPT Presentation
Chemical properties that a fg ect binding of enzyme-inhibiting drugs to enzymes Research proposal by Dan Nacu ? Global Pharmaceutical Industry $400,000,000,000 http:/ /www.who.int/trade/glossary/story073/en/ Drug Development Computer
SLIDE 2
SLIDE 3
http:/ /www.who.int/trade/glossary/story073/en/
$400,000,000,000
Global Pharmaceutical Industry
SLIDE 4
Computer Drug Development Simulations
SLIDE 5
How can it be done?
Simulation Models
Shape Complementarity Chemical Properties
SLIDE 6
How can it be done?
Chemical Properties
Solvent Accessible Surface Area Hydrophobicity Electrostatics Van Der Waal’s Forces Residue Pair potential Desolvation Energies Atomic Contact Energies Complementary Determining Regions etc… A lot of options…
SLIDE 7
Its been done before… in a difgerent way.
Li et al, 2007
Their Equation
SLIDE 8
Its been done before…in a difgerent way.
Li et al, 2007
Their Results
SLIDE 9
How will this be difgerent?
Introducing HINT
Hydropathic INTeractions
The HINT Equation
SLIDE 10
Why HINT?
Its more true to life
SLIDE 11
HINT
two atoms i j
SLIDE 12
HINT
affjnity score for single interaction
All interactions are summed at end
SLIDE 13
HINT
hydrophobic atom constant
i
hydrophobic character
- f interaction
hydrophobic atom constant
j
sum
Log P 1-octanol/water
SLIDE 14
HINT
accessible (inside) non-accessible
SLIDE 15
HINT
Electrostatics
returns +1 or -1
hydrophobic-hydrophobic hydrophobic-polar acid-base hydrogen bond polar-polar acid-acid base-base
SLIDE 16
HINT
Atomic Distance i j j j
decreased interaction
SLIDE 17
HINT
Lennard-Jones Potential i j i j i j
too close too far just right
SLIDE 18
What can be done?
By weighing each variable in HINT, the most important chemical property for enzyme/ inhibitor complexes can be found.
Why exponents?
SLIDE 19
Why exponents?
Li et al HINT
SLIDE 20
Overview of Experiment
Benchmark 5 PDB File Modifjed Equation Similarity to True-Structure Decision
SLIDE 21
What’s the experiment?
Start with 46 enzyme inhibitor complexes from the Benchmark 5.
SLIDE 22
What’s the experiment?
Bound Unbound
FTDock
What’s FTDock?
SLIDE 23
FTDock
Rotation & translation in 3D space to fjnd all possible confjgurations.
SLIDE 24
What’s the experiment?
FTDock
Huge list of possible complexes
SLIDE 25
What’s the experiment?
Ligand_Root-Mean-Square-Deviation
Huge list of possible complexes
L_RMSD Testing
What’s L_RMSD Testing?
SLIDE 26
L_RMSD Testing
atom x1 - atom x2 ( )2
Σ
total # of atoms
SLIDE 27
What’s the experiment?
L_RMSD Testing
Top 20 Structures
SLIDE 28
What’s the experiment?
Top 20 Structures
For 46 complexes = 920 simulated structures. For both bound and unbound, 1,840 total
Lets look at just one
SLIDE 29
What’s the experiment?
Why these exponents?
SLIDE 30
What’s the experiment?
23,000 HINT Scores for Bound 23,000 HINT Scores for Unbound
In the end…
46•20•5•5 = 23,000
6,325,000 scores
SLIDE 31
What’s the experiment?
L_RMSD Testing
Find best match for each complex
Highest 50 HINT Scores for each complex
SLIDE 32
Interpreting result
bij = ai aj Si Sj (Tij)1.5 Rij + rij
Best L_RMSD Score: 4 Å HINT weight used: Enzyme-Inhibitor Simulated Complex Electrostatics
SLIDE 33
Possible Results
SLIDE 34
Possible Results
SLIDE 35
In the future…
Difgerent models (besides HINT) Difgerent complexes (besides enzyme/inhibitor)
SLIDE 36
Questions?
SLIDE 37
Extras
SLIDE 38
Electrostatics Table
Kellogg and Abraham, 2001