tools for rational drug design
TRANSCRIPT
![Page 1: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/1.jpg)
Tools for rational drug design
E. MikrosUniversity of Athens
![Page 2: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/2.jpg)
Drug Discovery & DevelopmentIdentify disease
Isolate proteininvolved in disease (2-5 years)
Find a drug effectiveagainst disease protein(2-5 years)
Preclinical testing(1-3 years)
Formulation
Human clinical trials(2-10 years)
Scale-up
FDA approval(2-3 years)
File
IND
File
NDA
• Discovery of new targets– target discovery,– target identification and
validation– proteomics,
• Discovery of new therapeutic molecules
– lead discovery, – lead identification and
optimization
![Page 3: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/3.jpg)
Technology is impacting this process
Identify disease
Isolate protein
Find drug
Preclinical testing
GENOMICS, PROTEOMICS & BIOPHARM.
HIGH THROUGHPUT SCREENING
MOLECULAR MODELING
VIRTUAL SCREENING
COMBINATORIAL CHEMISTRY
IN VITRO & IN SILICO ADME MODELS
Potentially producing many more targetsand “personalized” targets
Screening up to 100,000 compounds aday for activity against a target protein
Using a computer topredict activity
Rapidly producing vast numbersof compounds
Computer graphics & models help improve activity
Tissue and computer models begin to replace animal testing
![Page 4: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/4.jpg)
In vitro, in vivo ……..in the market
In silico ?
• Gene expression analysis• Prediction of gene function• Protein structure• Virtual screening• ADME prediction
“….it is impossible to make every molecule that could be made …..”
![Page 5: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/5.jpg)
Structure-Based Drug DesignO-5
C-1
C-2C-3
C-4
C-5
O
O
O
H
C-6H
H
H
OH
O-1
O
C-1'
C-2'C-3'
C-4'
C-5'OH
OH O
C-6'
OH
CH3
H
ΦΨ
O-5
C-1
C-2C-3
C-4
C-5
O
O
O
H
C-6H
H
H
OH
O-1
O
C-1'
C-2'C-3'
C-4'
C-5'OH
OH O
C-6'
OH
CH3
H
ΦΨ
NHHO
O
O
O
OH
HO HN
HO
O
NHHO
O
O
O
OH
HO HN
HO
O
NHHO
O
O
O
OH
OH HN
O
OH
NHHO
O
O
O
OH
HO HN
HO
ONHHO
O
O
O
OH
OH HN
HO
O
O
O
NHHO
O
O
O
OH
OH HN
HO
O
OO
OO
NHHO
O
O
O
OH
OH HN
HO
O
ON
O N
NHHO
O
O
O
OH
HO HN
HO
O
NHHO
O
O
O
OH
OH HN
O
OHNHHO
O
O
O
OH
OH HN
HO
O
O
O
NHHO
O
O
O
OH
OH HN
HO
O
OO
OO
NHHO
O
O
O
OH
OH HN
HO
O
ON
O N
NHHO
O
O
O
OH
HO HN
HO
O
NHHO
O
O
O
OH
OH HN
O
OH
NHHO
O
O
O
OH
HO HN
HO
O
NHHO
O
O
O
OH
OH HN
HO
O
O
O
NHHO
O
O
O
OH
OH HN
HO
O
OO
OO
NHHO
O
O
O
OH
OH HN
HO
O
ON
O N
NHHO
O
O
O
OH
HO HN
HO
O
NHHO
O
O
O
OH
HO HN
HO
O
Virtual screening
Receptor Structure Chemical Library
Lead compound
![Page 6: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/6.jpg)
Protein Structure
![Page 7: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/7.jpg)
Protein StructureNMR spectroscopy
![Page 8: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/8.jpg)
Virtual screening
Prediction of Receptor Ligand Interactions
![Page 9: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/9.jpg)
Calculation of the energy
• Quantum Mechanics– ab initio – Semi-empirical
• Molecular Mechanics
Simulation of the Structure
![Page 10: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/10.jpg)
Molecular Mechanics
Energy is a function of the coordinates. Coordinates are function of the energy.
![Page 11: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/11.jpg)
( )
( )
( )[ ]
∑
∑
∑
∑
∑
+
−
+
−+
−+
−=
nonbondedji ij
ji
nonbondedji ij
ij
ij
ijij
torsionsall
anglesall
bondsallb
rqq
rR
rR
nK
K
bbKU
, 0
,
612
20
20
4
2
cos1
21
21
επε
ε
φ
θθ
φ
θ
Potential Energy Function
![Page 12: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/12.jpg)
Energy Minimization
“evolution” can be performed by systematic variation of the atom positions towards the lower energy directions. This procedure is called
“structure optimization” or “energy minimization”
![Page 13: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/13.jpg)
Conformational Search
• Systematic Search• Molecular Dynamics
• Simulated Annealing• Monte Carlo
exploring the energy landscape
![Page 14: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/14.jpg)
Conformations
• Equilibrium between conformers.
• Conformational Space• ALL possible
conformations undergiven constraints.
−−=
TkAEBEP
B
)()(exp
![Page 15: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/15.jpg)
![Page 16: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/16.jpg)
![Page 17: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/17.jpg)
![Page 18: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/18.jpg)
Systematic SearchO-Glucosides C-Glucosides
One rotatable bond Energy profile
![Page 19: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/19.jpg)
11.664 structures
C-Cellobiose
Relaxed map O-Cellobiose
O-5
C-1
C-2C-3
C-4
C-5
O
O
O
H
C-6H
H
H
OH
O-1
O
C-1'
C-2'C-3'
C-4'
C-5'OH
OH O
C-6'
OH
CH3
H
ΦΨ
![Page 20: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/20.jpg)
Monte Carlo 10.000 steps15 Degrees of Freedom3500 Structures up to 50 Kcal/mol
Quercetin-3-(α-L-Rha-2-1-α-L-Ara)
![Page 21: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/21.jpg)
![Page 22: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/22.jpg)
Molecular Dynamics
kTmv23
21 2 =
Molecular dynamicsuses thermal energyto explore the energysurface
position
Energy
![Page 23: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/23.jpg)
Receptor-Ligand Interactions
• Docking Calculations – Molecular Dynamics– Monte Carlo– Genetic Algorithms– Combinatorial Docking– Low Mode Search
![Page 24: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/24.jpg)
![Page 25: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/25.jpg)
Free Energy
]][[][
1
1
LPC
kkKbind ==
−
bindbind KRTΔG ln−=
ligand
protein
complex
water
polar and
non-polar
interactions with the solvent
polar and
non-polar
protein-ligand interactions
entropic effects
k1 k-1
![Page 26: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/26.jpg)
Virtual Screening
Glide Software• Application of a series of
filters that rapidly funneldown the possible ligandpositions and orientationsto a manageable numberfor detailed examination.
Halgren TA t al, J Med Chem. 2004, 47, 1750
![Page 27: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/27.jpg)
Συνάρτηση βαθμολόγησης GLIDE
•• ΗμιεμπειρικήΗμιεμπειρική συνάρτησησυνάρτηση αξιολόγησηςαξιολόγησης GLIDESCOREGLIDESCORE
•• ΑθροισμαΑθροισμα όρωνόρων πουπου περιγράφουνπεριγράφουν αλληλεπιδράσειςαλληλεπιδράσεις μεταξύμεταξύπρωτεΪνηςπρωτεΪνης, , μικρομορίουμικρομορίου καικαι διαλύτηδιαλύτη
•• ΕκτίμησηΕκτίμηση συγγένειαςσυγγένειας πρόσδεσηςπρόσδεσης ΔΔGGπρόσδεσηςπρόσδεσης
![Page 28: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/28.jpg)
![Page 29: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/29.jpg)
![Page 30: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/30.jpg)
![Page 31: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/31.jpg)
3D STRUCTURES3D STRUCTURES
STEP 1: DOCKINGSTEP 1: DOCKINGEVALUATION OF THE BINDING GEOMETRIES (RELATIVE ORIENTATION) OF EACH COMPOUND TO THE TARGET MACROMOLECULE
•STOCHASTIC MONTE CARLO SAMPLING
•RIGID REPRESENTATION OF THE PROTEIN BASED ON GRID CALCULATIONS
•FLEXIBLE REPRESENTATION OF THE LIGAND
STEP 2: SCORINGSTEP 2: SCORINGEVALUATION OF THE BINDING AFFINITY OF EACH COMPOUND
•ΔG APPROXIMATED BY SEMIEMPIRICAL SCORING FUNCTION:
RANKING OF THE RANKING OF THE COMPOUNDS COMPOUNDS
VIRTUAL SCREENINGVIRTUAL SCREENINGLIBRARY COMPOUNDSLIBRARY COMPOUNDS
GLIDE v.3GLIDE v.3SCHRSCHRÖÖDINGER Inc.DINGER Inc.
![Page 32: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/32.jpg)
COMPOUND COMPOUND LIBRARYLIBRARY
RANDOM RANDOM SCREENINGSCREENING
VIRTUAL VIRTUAL SCREENINGSCREENING
BIOLOGICAL BIOLOGICAL EVALUATIONS:EVALUATIONS:
13641364MOLECULESMOLECULES
7070MOLECULESMOLECULES
ACTIVES ACTIVES RECOVERED:RECOVERED:
1717ACTIVESACTIVES
1010ACTIVESACTIVES
Virtual Screening
![Page 33: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/33.jpg)
LEAD LEAD COMPOUNDCOMPOUND
DRUG DRUG DEVELOPMENTDEVELOPMENT
RECEPTOR BASED DESIGNRECEPTOR BASED DESIGN••CRYSTALLOGRAPHY, NMRCRYSTALLOGRAPHY, NMR••DOCKING SIMULATIONSDOCKING SIMULATIONSOR OR LIGAND BASED DESIGNLIGAND BASED DESIGN••QSAR, SIMILARITYQSAR, SIMILARITY
SYNTHESISSYNTHESIS
BIOLOGICAL EVALUATIONBIOLOGICAL EVALUATION••IN VITRO ASSAYSIN VITRO ASSAYS••IN VIVO TESTSIN VIVO TESTS
LEAD OPTIMIZATIONLEAD OPTIMIZATION••DOCKING SIMULATIONSDOCKING SIMULATIONS••STRUCTURESTRUCTURE--ACTIVITY RELATIONSACTIVITY RELATIONS•• ADMEADME
VIRTUAL OR VIRTUAL OR PHYSICAL PHYSICAL SCREENINGSCREENING
DRUGDRUG
COMPUTATIONAL COMPUTATIONAL CHEMISTRY DOMAINCHEMISTRY DOMAIN
![Page 34: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/34.jpg)
Drug likeness• Enriching screening libraries with drug-like
compounds• “fail fast, fail cheap” strategy• Manual classification is time-consuming and bias• Computational approaches speeds up the
screening, reduce the size and improves the quality of combinatorial libraries
• Assumption: typical drugs have something in common that other compounds lack
![Page 35: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/35.jpg)
Cheminformatics
• Lipinski Rule of Five• Poor absorption and permeation are more likely to occur
when there are – more than 5 hydrogen-bond donors, – more than 10 hydrogen-bond acceptors, – the molecular mass is greater than 500, – the log P value is greater than 5.
Lipinski et al., Adv. Drug Deliv. Rev. 23, 3-25 (1997)
![Page 36: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/36.jpg)
Ligand Based Drug Design
• Physicochemical Properties– H-bond donor-acceptor, ClogP, logD, charge
distribution, pK
• Pharmacophore concept
![Page 37: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/37.jpg)
Binary Fingerprints
![Page 38: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/38.jpg)
The pharmacophore feature pair histograms of captopril, and thepharmacophore fingerprint of captopril.
Chemaxon Pharmacophore fingerprints
![Page 39: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/39.jpg)
Similarity and Dissimilarity
![Page 40: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/40.jpg)
![Page 41: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/41.jpg)
RESULTS: VIRTUAL SCREENING
NCI Repository: 260.000 compounds
VS
33--DIMENSIONAL MOLECULAR SIMILARITYDIMENSIONAL MOLECULAR SIMILARITY
22--DIMENSIONAL FINGERPRINTSDIMENSIONAL FINGERPRINTS
DOCKINGDOCKING
![Page 42: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/42.jpg)
RESULTS: VIRTUAL SCREENING
NCI Repository: 260.000 compounds
VS 33--DIMENSIONAL MOLECULAR SIMILARITYDIMENSIONAL MOLECULAR SIMILARITY
![Page 43: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/43.jpg)
RESULTS: VIRTUAL SCREENING
NCI Repository: 260.000 compounds
VS 22--DIMENSIONAL FINGERPRINTSDIMENSIONAL FINGERPRINTS
![Page 44: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/44.jpg)
RESULTS: VIRTUAL SCREENING
NCI Repository: 260.000 compounds
VS DOCKINGDOCKING
![Page 45: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/45.jpg)
260.000
RESULTS: VIRTUAL SCREENING
40
VS
Consensus of 3 differ ent screeni ng me t hodol ogi es Consensus of 3 differ ent screeni ng me t hodol ogi es Selected compounds for Selected compounds for
in vitro screening:in vitro screening:
0.00015% of total collec on 0.00015% of total collec on
![Page 46: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/46.jpg)
Metabonomicsgenome
proteome metabolome
![Page 47: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/47.jpg)
Metabonomics
Quantitative measurement of multivariate metabolicresponses of multicellular systems to pathophysiological stimuli or genetic
modification
![Page 48: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/48.jpg)
1H NMR Spectrum of Untreated Human Urine
Metabonomics
![Page 49: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/49.jpg)
A procedure for Metabonomics
DSS
PCA
Tissue or biofluidsample
Measure themetabolite profile
Treat metabolite profile asstatistical ‘object’ for
classification purposes
Explore metaboliteprofile to gain
mechanistic insight
normal
PKU
MSUDApplication of NMR spectroscopy combined with principal component analysis in detecting inborn errors of metabolism using blood spots. A metabonomic approachM.A. Constantinou, E. Papakonstantinou, M. Spraul, K. Shulpis, M.A. Koupparis, E. MikrosAnalytica Chimica Acta, 511, 303-312, 2004
![Page 50: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/50.jpg)
x2
x3
x1
PC1
PC2
n,l=data points; n,l = projection
•Two PCs make a plane (window) in the K-dimensional variable space. The points are projected down onto theplane which is lifted out and viewed as a two dimensional plot. •This is the scores plot
•similarities or differences between samples can now be seen.
•A corresponding loading plot describes the variables relationships
•allows interpretation of the scores plot by showing which variables are responsible for similarities and differences between samples.
PCA theory PCA theory –– step by stepstep by step
![Page 51: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/51.jpg)
![Page 52: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/52.jpg)
The Loadings PlotsThe Loadings Plots
Loading (p): described the variation in the variable direction i.e. similarity/ dissimilarity between variables, and also explains the variation in scores. The loading (p) describes the original variables importance for respective PC. This is the same as the similarity in direction between the original variable and the PC.
The loading (p) is described as the cosine of the angle between the original variable and the PC.
PCPC22
PCPC11
Projection of (Projection of (rtrtxx, , m/zm/zxx))
px,1
px,2X1sample i θ1,2
θ2,2
PCPC22
p = cos θ∑ pi
2 = 1
X2
With px,1 = cos (θx,1) and px,2 = cos (θx,2) and θx,1 : angle between axe (rtx, m/zx) and PC1and θx,2 : angle between axe (rtx, m/zx) and PC2
ti1
ti = pj xi,j
![Page 53: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/53.jpg)
![Page 54: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/54.jpg)
y1
y2
u
0
01
1
t
u
x1
x2
x3
t(pca)t(pls-da)
PLSPLS--DADA
Metabonomic space
XXDescribes variation in NMR dataDescribes variation in NMR data
Dicriminant space
YYDefining the known classesDefining the known classes
‘‘Inner relationInner relation’’Maximising correlation between t and uMaximising correlation between t and u
Partial Least Squares or Projection to latent structure.
![Page 55: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/55.jpg)
Models both the X & Y matrices simultaneously to find the latentvariables in x that will predict the latent variables in Y the best.These PLS-Components are similar to principal components and willalso be referred to as PCs.
Partial Least Squares or Projection to latent structure.Partial least squares (PLS) is a method for constructing predictive models whenthe factors are many and highly collinear.
![Page 56: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/56.jpg)
Rapid and noninvasive diagnosis of the presence and severity of coronary heart disease using 1H-NMR-based metabonomicsJoanne T. Brindle, Henrik Antti, Elaine Holmes, George Tranter, Jeremy K. Nicholson, Hugh W.L. Bethell, Sarah Clarke, Peter M. Schofield, Elaine McKilligin, David E. Mosedale & David J. Grainger Nature Medicine 8, 1439 - 1445 (2002)
Clinical Diagnosis
Normal coronaryArteries (n=30)
SevereAtherosclerosis(n=36)
Predicting Coronary Artery Disease In Humans
![Page 57: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/57.jpg)
Predicting Coronary Artery Disease In Humans
![Page 58: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/58.jpg)
gastric and colon cancer
-10
0
10
-20 -10 0 10 20t[2
]t[1]
GASTRICCOLON
-10
0
10
-20 -10 0 10 20
t[2]
t[1]
CTL
GASTRICCOLON
PCA
![Page 59: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/59.jpg)
-20
-10
0
10
20
t[2]
t[1]
2010 30-20 -10-30
colorcolor
region
region
0
Greek Wines classification
1H NMR-Based Metabonomics for the Classification of Greek Wines According to Variety, Region and Vintage – Comparison with HPLC Data. Anastasiadi, M;J. Agr. Food. Chem. (2009); 57; 11067-11074
![Page 60: Tools for rational drug design](https://reader033.vdocuments.net/reader033/viewer/2022051522/589da32c1a28abae498bcf75/html5/thumbnails/60.jpg)