1. Computer-Aided Drug Discovery
Chemical Data and
Computer-Aided Drug Discovery
Mike Gilson School of Pharmacy
2-0622

2. Outline Overview of drug discovery
Structure-based computational methods
When we know the structure of the targeted protein
Ligand-based computational methods
When we don’t know the protein’s structure

3. What is a drug?

4. Small Molecule Drugs Aspirin Taxol Darunavir Sildenafil (Viagra)
Glipizide (Glucotrol)
Must be FDA-approved
Digoxin

5. Nanoparticles (e.g., packaged small-molecule drugs)
Doxil (liposome package,
extended circulation time,milder toxicity)
Abraxane (albumin-packaged taxol)

6. Biopharmaceuticals
Erythropoietin slightly engineered. Enbrel an engineered construct from parts of two natural proteins.
Erythropoietin (EPO) Stabilized variant of a natural protein hormone
Etanercept (Enbrel) Protein with TNF receptor + Ab Fc domain
Scavenges TNF, diminishes inflammation

7. How are drugs discovered?

8. Natural Products Aspirin Digoxin Taxol Pacific Yew Willow Foxglove
Natural products, initially utterly empirical.
Penicillin, chinese remedies, taxol etc.
Pacific Yew
Willow
Foxglove

9. How Aspirin Works Aspirin inflammation platelet activation
retrospective elucidation of natural product mechanism.
inflammation
platelet activation
platelet inactivation

10. Biomolecular Pathways and Target Selection E.g. signaling pathways
Target protein
Reductionist approach

11. Empirical Path to Ligand Discovery
Compound library (commercial, in-house, synthetic, natural)
High throughput screening (HTS)
Hit confirmation
Lead compounds (e.g., µM Kd)
Lead optimization
(Medicinal chemistry)
Animal and clinical evaluation
Potent drug candidates (nM Kd)

12. Compound Libraries Government (NIH) Commercial (also in-house pharma)
Academia

13. Computer-Aided Ligand Design
Aims to reduce number of compounds synthesized and assayed
Lower costs
Less chemical waste
Faster progress

14. Structure of Targeted Protein Known: Structure-Based Drug Discovery
Scenario 1
Structure of Targeted Protein Known: Structure-Based Drug Discovery
HIV Protease/KNI-272 complex

15. - Protein-Ligand Docking Structure-Based Ligand Design
Docking software Search for structure of lowest energy
Potential function Energy as function of structure
VDW - +
Screened Coulombic
Minimal words here.
Dihedral

16. Energy Determines Probability (Stability) Boltzmann distributionx

17. Structure-Based Virtual Screening
Compound database
3D structure of target (crystallography, NMR, modeling)
Virtual screening (e.g., computational docking)
Candidate ligands
Ligand optimization Med chem, crystallography, modeling
Experimental assay
Discuss optimization. Solve structure, see how it actually binds, make new designs, synthesize and retest
Ligands
Drug candidates

18. Fragmental Structure-Based Screening
“Fragment” library
3D structure of target (crystallography, NMR, modeling)
Fragment docking
Compound design
Experimental assay and ligand optimization Med chem, crystallography, modeling
Drug candidates

19. Potential Functions for Structure-Based Design
Energy as a function of structure
Physics-Based
Knowledge-Based

20. Physics-Based Potentials Energy terms from physical theory
Van der Waals interactions (shape fitting)
Bonded interactions (shape and flexibility)
Coulombic interactions (charge-charge complementarity)
Hydrogen-bonding

21. Common Simplifications Used in Physics-Based Docking
Quantum effects approximated classically
Protein often held rigid
Configurational entropy neglected
Influence of water treated crudely
Fitting is often used.

22. Proteins and Ligand are Flexible+
Ligand
Protein
Complex
DGo

23. Binding Energy and Entropy
EFree
Unbound states
EBound
Bound states
Energy part
Entropy part

24. Structure-Based Discovery
Physics-oriented approaches
Weaknesses
Fully physical detail becomes computationally intractable
Approximations are unavoidable
Parameterization still required
Strengths
Interpretable, provides guides to design
Broadly applicable, in principle at least
Clear pathways to improving accuracy
Status
Useful, far from perfect
Multiple groups working on fewer, better approxs
Force fields, quantum
Flexibility, entropy
Water effects
Moore’s law: hardware improving

25. Knowledge-Based Docking Potentials
Histidine
Ligand carboxylate
Aromatic stacking

26. Probability Energy Boltzmann: Inverse Boltzmann:
Example: ligand carboxylate O to protein histidine N
Find all protein-ligand structures in the PDB with a ligand carboxylate O
For each structure, histogram the distances from O to every histidine N
Sum the histograms over all structures to obtain p(rO-N)
Compute E(rO-N) from p(rO-N)

27. Knowledge-Based Docking Potentials
“PMF”, Muegge & Martin, J. Med. Chem. 42:791, 1999
A few types of atom pairs, out of several hundred total
Nitrogen+/Oxygen-
Aromatic carbons
Aliphatic carbons
Atom-atom distance (Angstroms)

28. Structure-Based Discovery
Knowledge-based potentials
Weaknesses
Accuracy limited by availability of data
Accuracy may also be limited by overall approach
Strengths
Relatively easy to implement
Computationally fast
Status
Useful, far from perfect
May be at point of diminishing returns

29. Limitations of Knowledge-Based Potentials
1. Statistical limitations (e.g., to pairwise potentials)
2. Even if we had infinite statistics, would the results be accurate? (Is inverse Boltzmann quite right? Where is entropy?)
100 bins for a histogram of O-N & O-C distances r1 r2
r10 …
10 bins for a histogram of O-N distances
rO-C
rO-N
rO-N

30. Structure of Targeted Protein Unknown: Ligand-Based Drug Discovery
Scenario 2
Structure of Targeted Protein Unknown: Ligand-Based Drug Discovery
e.g. MAP Kinase Inhibitors
Using knowledge of existing inhibitors to discover more

31. Why Look for Another Ligand if You Already Have Some?
Experimental screening generated some ligands, but they don’t bind tightly
A company wants to work around another company’s chemical patents
An high-affinyt ligand is toxic, is not well-absorbed, etc.

32. Ligand-Based Virtual Screening
Compound Library
Known Ligands
Molecular similarity
Machine-learning
Etc.
Candidate ligands
Optimization Med chem, crystallography, modeling
Assay
Actives
Potent drug candidates

33. Sources of Data on Known Ligand Journals, e.g., J. Med. Chem.

34. Some Binding and Chemical Activity Databases
PubChem (NIH) pubchem.ncbi.nlm.nih.gov
ChEMBL (EMBL)
BindingDB (UCSD)

35. BindingDB

36. Finding Protein-Ligand Data in BindingDB
e.g., by Name of Protein “Target”
e.g., by Ligand Draw  Search

37. Sample Query Results BindingDB to PDB

38. PDB to BindingDB

39. Sample Query Results
Download data in machine-readable format

40. Machine-Readable Chemical Format
Structure-Data File (SDF)
SDF Format Defines Chemical Bonds
PDB Format Lacks Chemical Bonding

41. There are Many Other Chemical File Formats Interconvert with Babel

42. Chemical Similarity Ligand-Based Drug-Discovery
Compounds (available/synthesizable)
Similar
Compare with known ligands
Test experimentally
Different
Don’t bother

43. Chemical Fingerprints Binary Structure Keys
Molecule 1
Molecule 2
phenyl
methyl
ketone
carboxylate
amide
aldehyde
chlorine
fluorine
ethyl
naphthyl
S-S bond
alcohol …

44. Chemical Similarity from Fingerprints Tanimoto Similarity or Jaccard Index, T
Intersection
NU=8
Union
Molecule 1
Molecule 2

45. Hashed Chemical Fingerprints
Based upon paths in the chemical graph
1-atom paths: C F N H S O
2-atom paths: F-C C-C C-N C-S S-O C-H
3-atom paths: F-C-C C-C-N C-N-H C-S-O
Each path sets a pseudo-random bit-pattern in a very long molecular fingerprint C
S-O
etc.

46. Maximum Common Substructure
Ncommon=34

47. Potential Drawbacks of Plain Chemical Similarity
May miss good ligands by being overly conservative
Too much weight on irrelevant details

48. Scaffold Hopping
Identification of synthetic statins by scaffold hopping
Zhao, Drug Discovery Today 12:149, 2007

49. Abstraction and Identification of Relevant Compound Features
Ligand shape
Pharmacophore models
Chemical descriptors
Statistics and machine learning

50. Φάρμακο (drug) + Φορά (carry)
Pharmacophore Models
Φάρμακο (drug) + Φορά (carry)
+ 1
Bulky hydrophobe
Aromatic
5.0 ±0.3 Å
3.2 ±0.4 Å
2.8 ±0.3 Å
A 3-point pharmacophore

51. Molecular Descriptors More abstract than chemical fingerprints
Physical descriptors
molecular weight
charge
dipole moment
number of H-bond donors/acceptors
number of rotatable bonds
hydrophobicity (log P and clogP)
Topological
branching index
measures of linearity vs interconnectedness
Etc. etc.
Rotatable bonds

52. A High-Dimensional “Chemical Space”
Each compound is at a point in an n-dimensional space
Compounds with similar properties are near each other
Descriptor 3
Descriptor 1
Descriptor 2
Point representing a compound in descriptor space

53. Statistics and Machine Learning
Some examples
Partial least squares
Support vector machines
Genetic algorithms for descriptor-selection

54. Summary Overview of drug discovery Computer-aided methods
Structure-based
Ligand-based
Interaction potentials
Physics-based
Knowledge-based (data driven)
Ligand-protein databases, machine-readable chemical formats
Ligand similarity and beyond
Mike Gilson, School of Pharmacy,

55. Activities and Discussion Topics
BindingDB: Advil Machine-readable format, Binding activities
PDB/BindingDB
2ONY at PDB BindingDB Substructure search Related data
 Similarity search
Combined computational approaches
(physics + knowledge)-based docking potentials
(ligand + structure)-based computational discovery
Other data-driven methods where it may be hard to get enough statistics
Validation of computational methods
Protein-ligand databases: getting data and assessing data quality

57. Drug Discovery Pipeline (One Model)
Target identification
Target validation
Assay development
Lead compound
(ligand) discovery
Lead optimization
Animal Pharmacokinetics, Toxicity
Target selection: may use pathway information and modeling
Target validation: e.g., knockouts, silencing RNA
Assay development: format, throughput, depending on how indent to discover compounds
Lead compound discovery
Phase I Clinical (safety, metab, PK)
Phase II Clinical (efficacy)
Phase III Clinical (comparison with existing therapy)

58. Updated Knowledge-Based PMF Potential
Muegge J. Med. Chem. 49: 5895, 2006