azevedolab.net Biologia de Sistemas … › resources › Univates_2018_03_22...• Apresentação • Biologia de Sistemas Computacional • Docking Molecular • CDK • Projeto

"Biologia de Sistemas Computacional Aplicada ao Desenvolvimento de Fármacos. Um Estudo de

Inibidores da Quinase Dependente de Ciclina (CDK)"

Prof. Dr. Walter F. de Azevedo Jr. Laboratório de Biologia de Sistemas Computacional

Escola de Ciências-Pontifícia Universidade Católica do Rio Grande do Sul-PUCRS

print(cdk)

azevedolab.net

• Apresentação

• Biologia de Sistemas Computacional

• Docking Molecular

• CDK

• Projeto SAnDReS

• Espaço de Funções Escores

• Outros Projetos em Desenvolvimento

• Usos Recentes do SAnDReS

• Considerações Finais

• Agradecimentos

Sumário

azevedolab.net

Prof. Harold Morowitz.

Santa Fé Institute. Disponível em:

http://www.santafe.edu/about/people/profile/Harold%20Moro

witz . Acesso em: 23 de março de 2018.

Biologia de Sistemas Computacional

“Computers are to biology what mathematics is to physics.”

azevedolab.net

http://www.santafe.edu/about/people/profile/Harold Morowitz













































Biologia Física

Biologia de

Sistemas

Química

Computação

Matemática


azevedolab.net

4


Genes mRNAs Proteínas Metabólitos

Genômica Transcriptômica Proteômica Metabolômica

Verificação

experimental

Análise computacional dos dados

Modelo

Computacional Previsão

Sistema Biológico

azevedolab.net

P

L

Docking Molecular

azevedolab.net

Algoritmos Evolucionários

azevedolab.net

Indivíduo

População


azevedolab.net

População da primeira geração

População da segunda geração

Fitness

x

y


azevedolab.net

101101011010 000010000100 000001111111 100111100111 111110011000 000001011010 000010111100 110100111100

Strings binárias

(populatção incial)

Operador seleção

Operador crossover

Operador mutação

110100111010 101101011100 100110011000

1111111100111 101101011010 100111100111 111110011000 110100111100

Strings binárias

(Nova população)


azevedolab.net

110100111010 101101011100

100110011000

1111111100111

110100111100 101101011010

100111100111 111110011000

Locus

Locus

Pais Descendentes

Crossover


azevedolab.net

110100111010 101101011100

100110011000

1111111100111

110100111100 101101011010

100111100111 111110011000

Locus

Locus

Pais Descendentes

Crossover

Gera um número aleatório

(Rn)

Rn<=Pc?

Escolha novo par de pais

Sim

Não


azevedolab.net

Initialização

A B

C D


azevedolab.net

Simulação de Docking Molecular

Ligante (chave)

Sítio ativo (fechadura)

azevedolab.net

N

zzyyxx

RMSD

N

j

jposejcristaljposejcristaljposejcristal

1

2

,,

2

,,

2

,,


azevedolab.net


azevedolab.net

Pai 1 Pai 2

crossover

Filha


azevedolab.net

Filha

Filha mutada

Mutação


azevedolab.net

Simulação de Docking Molecular azevedolab.net

azevedolab.net

Algoritmos Evolucionários (Evolução Diferencial)

Operador seleção

Operador crossover

Operador mutação

Critério de parada

satisfeito?

Mostra resultados

Sim

Não

azevedolab.net

azevedolab.net

cromossomo(n) = cromossomo(m) + peso.[cromossomo(k) - cromossomo(l)]

o peso varia entre 0 e 2.


azevedolab.net

azevedolab.net


azevedolab.net

azevedolab.net

Cromossomo k Cromossomo l

Cromossomo m

cromossomo(n) = cromossomo(m) + peso.[cromossomo(k) - cromossomo(l)]


Cromossomo n

Cromossomo j

azevedolab.net

azevedolab.net

Cromossomo n

Cromossomo n mutado

Mutação


Cromossomo j é deletado e

substituído pelo n

azevedolab.net

azevedolab.net

CDK azevedolab.net

azevedolab.net

CDK

Referências: De Azevedo WF, Leclerc S, Meijer L, Havlicek L, Strnad M, Kim SH. Inhibition of cyclin-dependent kinases by purine analogues: crystal structure of human cdk2 complexed with roscovitine. Eur J Biochem. 1997; 243(1-2): 518-26. PubMed De Azevedo WF Jr, Mueller-Dieckmann HJ, Schulze-Gahmen U, Worland PJ, Sausville E, Kim SH. Structural basis for specificity and potency of a flavonoid inhibitor of human CDK2, a cell cycle kinase. Proc Natl Acad Sci U S A. 1996; 93(7): 2735-40. PubMed Kim SH, Schulze-Gahmen U, Brandsen J, de Azevedo Júnior WF. Structural basis for chemical inhibition of CDK2. Prog Cell Cycle Res; 2: 137-45. PubMed

DNA damage

p53 p21

CDK2+cyclin A

azevedolab.net

azevedolab.net

http://www.ncbi.nlm.nih.gov/pubmed/9030780



Thomsen R, Christensen MH. MolDock: a new technique for high-accuracy molecular docking. J Med Chem. 2006;49:3315– 21.

SAnDReS azevedolab.net

azevedolab.net

SAnDReS

+25 mil linhas de código Código aberto Testado em diversos sistemas biológicos

azevedolab.net

azevedolab.net

SAnDReS

www.sandres.net

azevedolab.net

azevedolab.net

8

SAnDReS

GUI window

Text window

azevedolab.net

azevedolab.net

SAnDReS

SAnDReS GUI

SAnDReS main

program

Input file (.in)

log files (.log)

csv files (.csv)

Plot files

Machine-learning

techniques

azevedolab.net

azevedolab.net

SAnDReS (Docking Molecular)

Binding Affinity Total Number of

Available Structures

Total Number of

Structures Determined

by X-ray

Crystallography

Total Number of

Structures

Determined by

NMR

Total Number of

Structures

Determined by

Neutron

Crystallography

Total Number of

Structures

Determined by

hybrid methods

Total Number of

Structures

Determined by

electronic

micrography

Percentage of

Structures Determined

by X-ray

Crystallography

Ki 5503 5467 34 1 1 0 99.35

Kd 6392 6044 344 1 1 2 94.56

Ka 110 110 0 0 0 0 100,0

IC50 6207 6176 29 1 1 0 99.50

DeltaG 139 136 2 0 1 0 97.84

DeltaH 59 59 0 0 0 0 100.0

azevedolab.net

azevedolab.net

Selection of

Biomolecular

System

Download

Structures and

Experimental

Binding

Information

Pre-Docking

Analysis

Re-Docking

Statistical

Analysis of Re-

docking Results

Ensemble-

Docking

Statistical

Analysis of

Ensemble-

Docking Results

Scoring Function

Calculation

Statistical

Analysis of

Scoring

Functions

Results

Virtual Screening

with a Dataset

with Decoys +

Actives

Statistical

Analysis of

Virtual Screening

with Decoys +

Actives

Virtual Screening

Simulation

Statistical

Analysis of

Virtual Screening

Simulation

Results

Best Hits


Referência:

Xavier MM, Heck GS, de Avila MB, Levin NM, Pintro VO, Carvalho NL, Azevedo WF Jr. SAnDReS a Computational Tool for Statistical

Analysis of Docking Results and Development of Scoring Functions. Comb Chem High Throughput Screen. 2016; 19(10): 801-812.

azevedolab.net

azevedolab.net

Selection of

Biomolecular

System

Download

Structures and

Experimental

Binding

Information

Pre-Docking

Analysis

Re-Docking

Statistical Analysis

of Re-Docking

Results

Ensemble-Docking


of Ensemble-

Docking Results

and Structural

Parameters

Scoring Function

Calculation

Building New

Scoring Functions

Results


of Scoring

Functions Results

Virtual Screening

with a Dataset with

Decoys + Actives


of Virtual

Screening with

Decoys + Actives

Virtual Screening

Simulation

Best Hits


of Virtual

Screening

Simulation Results


Referência:

Xavier MM, Heck GS, de Avila MB, Levin NM, Pintro VO, Carvalho NL, Azevedo WF Jr. SAnDReS a Computational Tool for Statistical

Analysis of Docking Results and Development of Scoring Functions. Comb Chem High Throughput Screen. 2016; 19(10): 801-812.

azevedolab.net

azevedolab.net

30

SAnDReS (CDK) (Re-docking) azevedolab.net

azevedolab.net

Referência: Levin NMB, Pintro VO, Bitencourt-Ferreira G, Mattos BB, Silvério AC, de Azevedo Jr. WF. Development of CDK-targeted scoring functions for prediction of binding affinity. Biophys Chem. 2018; 235: 1–8. https://doi.org/10.1016/j.bpc.2018.01.004 Link PubMed PDF

https://doi.org/10.1016/j.bpc.2018.01.004

https://www.sciencedirect.com/science/article/pii/S0301462217304258

https://www.ncbi.nlm.nih.gov/pubmed/29407904

https://www.researchgate.net/publication/322871814_Development_of_CDK-targeted_scoring_functions_for_prediction_of_binding_affinity

SAnDReS (CDK) (Re-docking)

Using MVD (Protocol 13)

Scoring Function/Energy Term RMSDa b p-value1c R2 d p-value2e

MolDock Score 1.383 0.880 3.741.10-17 0.853 1.368.10-21

Re-rank Score 0.846 0.359 1.048.10-2 0.033 2.037.10-1

Interaction Score 0.809 0.941 2.771.10-24 0.764 1.128.10-16

Protein Score 0.809 0.941 2.771.10-24 0.764 1.128.10-16

Internal Score 8.758 -0.261 6.667.10-2 0.026 2.642.10-1

H-Bond Score 1.670 0.576 1.181.10-5 0.460 6.254.10-8

LE1 Score 1.383 0.880 3.741.10-17 0.853 1.367.10-21

LE3 Score 0.846 0.359 1.048.10-2 0.033 2.037.10-1

Docking Score 0.600 0.899 7.047.10-19 0.871 5.248.10-23

Displaced Water Score 6.288 0.352 1.213.10-2 0.360 4.117.10-6

Using AD4


Free Energy 1.230 0.652 4.087.10-2 0.549 1.419.10-2

Final Intermolecular Energy 1.230 0.675 3.231.10-2 0.573 1.128.10-2

vdW+Hbond+desolv Energy 0.740 0.766 9.787.10-3 0.767 9.018.10-4

Electrostatic Energy 2.210 -0.612 6.023.10-2 0.793 5.532.10-4

Final Total Internal Energy 1.330 -0.317 3.720.10-1 0.028 6.437.10-1

Using Vina


Affinity 1.359 0.232 3.259.10-1 0.202 4.669.10-2

Gauss1 6.374 -0.179 4.503.10-1 0.114 1.454.10-1

Gauss2 5.770 0.090 7.052.10-1 0.059 3.001.10-1

Repulsion 6.928 -0.060 8.011.10-1 0.005 7.775.10-1

Hydrophobic 6.819 -0.495 2.658.10-2 0.043 3.821.10-1

Hydrogen 6.597 0.106 6.576.10-1 0.000 9.461.10-1

azevedolab.net

azevedolab.net

41

Type of CDK PDB Access Codes

CDK2 1DI8,1DM2,1E9H,1FVT,1FVV,1GII,1H00,1H07,1H0W,1H1Q,1H1R,1JVP,

1KE5,1KE6,1KE7,1KE8,1KE9,1OGU,1OI9,1OIQ,1OIR,1OIT,1OIU,1OIY,1P2A,1PXI,1PXL,1PYE,1R78,1URW,1V1K,1VYW,1VYZ,1W0X,

1WCC,1Y8Y,1Y91,1YKR,2A0C,2B52,2B53,2B54,2BHE,2BKZ,2BPM,2BTS,2C4G,2C5N,2C5O,2C5Y,2C68,2C69,2C6I,2C6K,2C6L,

2C6M,2DS1,2DUV,2G9X,2I40,2IW6,2IW9,2R3F,2R3G,2R3H,2R3I,2R3J,2R3K,2R3L,2R3M,2R3N,2R3O,2R3P,2R64,2UUE,2UZB,2UZD,2U

ZE,2UZL,2UZN,2UZO,2VTA,2VTH,2VTI,2VTJ,2VTL,2VTM,2VTN,2VTO,2VTP,2VTQ,

2VTR,2VTS,2VTT,2VU3,2VV9,2W05,2W06,2W17,2W1H,2WEV,2WIH,2WXV,3BHT,3BHU,3BHV,3DDP,3DDQ,3DOG,3EZR,3EZV,3FZ1

,3IG7,3IGG,3LE6,3LFN,3LFS,3NS9,3PJ8,3PXZ,3PY0,3QQK,3QTR,3QTS,3QTU,3QTW,3QTX,3QTZ,3QU0,3R8U,3R8V,3R8Z,3R9D,

3R9H,3R9N,3R9O,3RAH,3RAL,3RJC,3RK5,3RK7,3RK9,3RKB,3RMF,3RNI,

3RPR,3RPV,3RPY,3RZB,3S00,3S0O,3S1H,3S2P,3TI1,3TIY,3TIZ,3ULI,3UNJ,

3UNK,3WBL,4BGH,4CFN,4CFW,4ERW,4EZ3,4GCJ,4LYN

CDK5 1UNG,1UNH,3O0G,4AU8

CDK6 4AUA

CDK8 3RGF

CDK9 3BLR,3LQ5,3TN8

SAnDReS (CDK) (Ensemble Docking) azevedolab.net

azevedolab.net

SAnDReS (CDK) (Ensemble Docking)


MolDock Scoref 0.440 0.155 3.944.10-2 0.038 9.913.10-3

Re-rank Scoref 0.538 0.262 4.451.10-4 0.015 1.078.10-1

Interaction Scoref 0.592 0.094 2.123.10-1 0.013 1.276.10-1

Protein Scoref 0.592 0.100 1.880.10-1 0.013 1.319.10-1

Internal Scoref 0.311 -0.164 2.943.10-2 0.003 4.436.10-1

H-Bond Scoref 0.311 0.309 3.014.10-5 0.010 1.952.10-1

LE1 Scoref 0.311 0.323 1.223.10-5 0.011 1.590.10-1

LE3 Scoref 0.311 0.352 1.719.10-6 0.012 1.565.10-1

Docking Scoref 0.440 0.133 7.858.10-2 0.026 3.186.10-2

Displaced Water Scoref 0.145 0.400 3.795.10-8 0.014 1.123.10-1

Free Energyg 0.900 0.137 7.016.10-2 0.177 5.991.10-9

Final Intermolecular Energyg 2.840 0.003 9.661.10-1 0.104 1.302.10-5

vdW+Hbond+desolv Energyg 2.840 0.036 6.354.10-1 0.268 1.807.10-13

Electrostatic Energyg 1.820 -0.054 4.778.10-1 0.553 2.888.10-32

Final Total Internal Energyg 2.840 -0.235 1.703.10-3 0.001 7.306.10-1

Affinityh 0.271 0.372 3.591.10-7 0.282 3.599.10-14

Gauss1h 1.213 -0.251 7.732.10-4 0.061 9.409.10-4

Gauss2h 4.1383 -0.134 7.601.10-2 0.073 2.935.10-4

Repulsionh 4.1383 -0.321 1.444.10-5 0.002 5.559.10-1

Hydrophobich 7.260 -0.093 2.207.10-1 0.041 7.068.10-3

Hydrogenh 5.593 -0.251 7.641.10-4 0.042 6.626.10-3

Docking results for all structures in the CDK data set aRMSD is the RMSD for the lowest scoring function value, and its unit is Å.

b is Spearman rank-order correlation coefficient. cp-value1 is related to . dR2 is the squared Pearson correlation coefficient. ep-value2 is related to R2 . fUsing MVD (protocol 13). gUsing AD4. hUsing Vina.

azevedolab.net

azevedolab.net

47

SAnDReS (CDK) (Binding Affinity) azevedolab.net

azevedolab.net






52

SAnDReS (CDK) (Binding Affinity)

Scoring Function/Energy Term a p-value1b R2 c p-value2d

MolDock Scoree -0.092 2.259.10-1 0.097 2.697.10-5

Re-rank Scoree -0.207 5.798.10-3 0.079 1.643.10-4

Interaction Scoree 0.038 6.170.10-1 0.067 5.315.10-4

Cofactor Scoree 0.115 1.277.10-1 0.022 5.147.10-2

Protein Scoree 0.069 3.643.10-1 0.056 1.496.10-3

Water Scoree -0.094 2.127.10-1 0.013 1.345.10-1

Internal Scoree 0.201 7.347.10-3 0.062 8.585.10-4

Electro Scoree 0.105 1.642.10-1 0.000 9.698.10-1

Electro Long Scoree -0.186 1.369.10-2 0.017 8.408.10-2

H-Bond Scoree 0.104 1.675.10-1 0.018 7.992.10-2

LE1 Scoree -0.041 5.861.10-1 0.002 5.627.10-1

LE3 Score -0.078 3.049.10-1 0.001 7.117.10-1

Free Energyf 0.167 2.660.10-2 0.000 8.983.10-1

Final Intermolecular Energyf 0.142 5.983.10-2 0.000 8.977.10-1

vdW+Hbond+desolv Energyf 0.213 4.510.10-3 0.000 8.941.10-1

Electrostatic Energyf 0.026 7.317.10-1 0.000 8.940.10-1

Final Total Internal Energyf 0.312 2.433.10-5 0.000 9.203.10-1

Torsional Free Energyf -0.128 9.082.10-2 0.017 8.689.10-2

Affinityg 0.275 2.188.10-4 0.005 3.339.10-1

Gauss1g -0.220 3.320.10-3 0.023 4.641.10-2

Gauss2g -0.287 1.122.10-4 0.022 4.798.10-2

Repulsiong -0.199 8.012.10-3 0.054 1.923.10-3

Hydrophobicg -0.178 1.789.10-2 0.055 1.765.10-3

Hydrogeng -0.196 9.114.10-3 0.044 5.324.10-3

Correlation between scoring function/energy term and experimental binding affinity (log(IC50)) for all structures in the CDK data set. a is Spearman rank-order correlation coefficient. bp-value1 is related to . cR2 is the squared Pearson correlation coefficient. dp-value2 is related to R2 . eUsing MVD. fUsing AD4. gUsing Vina.

azevedolab.net

azevedolab.net

Referências: De Azevedo WF Jr, Dias R. Evaluation of ligand-binding affinity using polynomial empirical scoring functions. Bioorg Med Chem. 2008; 16(20):9378-82. Levin NMB, Pintro VO, Bitencourt-Ferreira G, Mattos BB, Silvério AC, de Azevedo Jr. WF. Development of CDK-targeted scoring functions for prediction of binding affinity. Biophys Chem. 2018; 235: 1–8. https://doi.org/10.1016/j.bpc.2018.01.004 Link PubMed PDF

Função Escore Polinomial

SAnDReS (CDK) (Machine Learning)

2

39

2

28

2

17326

31521433

22110score

xxxxx

xxxxx

xx

azevedolab.net

azevedolab.net





56

SAnDReS (CDK) (Machine Learning) azevedolab.net

azevedolab.net






52

Scoring Function/Energy Term Training Set Test Set

p-value1 p-value1

MolDock Scorea 0.059 5.179.10-1 -0.291 3.265.10-2

Re-rank Scorea -0.162 7.515.10-2 -0.132 3.405.10-1

Interaction Scorea 0.154 8.952.10-2 -0.195 1.568.10-1

Co-factor Scorea -0.010 9.170.10-1 0.265 5.243.10-2

Protein Scorea 0.211 1.943.10-2 -0.298 2.874.10-2

Water Scorea -0.130 1.550.10-1 0.024 8.614.10-1

Internal Scorea 0.170 6.150.10-2 0.252 6.632.10-2

Electro Scorea 0.041 6.549.10-1 -0.123 3.762.10-1

Electro Long Scorea -0.173 5.712.10-2 -0.060 6.667.10-1

H-Bond Scorea 0.187 3.932.10-2 0.027 8.483.10-1

LE1 Scorea -0.016 8.603.10-1 -0.026 8.500.10-1

LE3 Scorea -0.077 3.984.10-1 -0.084 5.463.10-1

Score482 a 0.390 9.065.10-6 0.346 1.044.10-2

Free Energyb 0.190 3.890.10-2 0.213 1.082.10-1

Final Intermolecular Energyb 0.200 2.961.10-2 0.172 1.975.10-1

vdW+Hbond+desolv Energyb 0.222 1.576.10-2 0.203 1.270.10-1

Electrostatic Energyb 0.047 6.106.10-1 -0.062 6.445.10-1

Final Total Internal Energyb 0.331 2.469.10-4 0.147 2.697.10-1

Torsional Free Energyb -0.176 5.659.10-2 0.043 7.494.10-1

Score281b 0.457 1.963.10-7 0.221 9.607.10-2

Affinityc 0.339 1.495.104 0.207 1.267.10-1

Gauss1c -0.297 1.006.10-3 -0.114 4.023.10-1

Gauss2c -0.347 1.025.10-4 -0.218 1.058.10-1

Repulsionc -0.173 5.819.10-2 -0.146 2.836.10-1

Hydrophobicc -0.148 1.061.10-1 -0.027 8.436.10-1

Hydrogenc -0.167 6.901.10-2 -0.235 8.080.10-2

Results for training set and test set for CDK data set. aUsing MVD. bUsing AD4. cUsing Vina.


azevedolab.net


x.y.0000030 +0.001136z-0.001529y +0.001829x - 7.074331- 482

score

where Re-rank (x), Internal (y), and Electro Long (z) Scores were used as explanatory variables. This polynomial equation

shows = 0.389 (p-value < 0.001) for the training set (122 structures) and = 0.345 (p-value = 0.0105) for a test set with 54

structures.

= 0.345 (p-value = 0.0105)

azevedolab.net

azevedolab.net







b.c.b+.a.b.a+.-.-score 200001600000840000016000001603477846281

where the explanatory variables were determined using AD4 taking a as the Free Energy, b as the Final Internal Energy, and c

as the Electrostatic Energy. This model shows = 0.457 (p-value < 0.001) for the training set (122 structures) and = 0.221 (p-

value = 0.096) for a test set with 54 structures.

= 0.457 (p-value < 0.001)

azevedolab.net

azevedolab.net






Lys 33

Asp 145

Glu 12

Glu 81

Asn 132

Asp 86 Gln 85

His 84

Lys 33 Glu 81

Asp 86

Asn 132

Asp 145

Residue Number


azevedolab.net






SAnDReS (CDK)

x.y.0000030 +0.001136z-0.001529y

+0.001829x - 7.074331- 482

score IC50


azevedolab.net

azevedolab.net





Exploring the Scoring Function

Heck GS, Pintro VO, Pereira RR, de Ávila MB, Levin NMB, de Azevedo WF. Supervised Machine Learning Methods Applied to Predict Ligand-Binding Affinity. Curr Med Chem. 2017; 24(23): 2459-2470.

azevedolab.net

azevedolab.net

Improvement of the predictive power of a scoring function generated with the

program SAnDReS. Figure created by Ms. Gabriela Bitencourt-Ferreira.

Exploring the Scoring Function azevedolab.net

azevedolab.net

Tool to Analyze the Binding Affinity Desenvolvido por: Amauri Duarte & Prof. Dr. Walter Filgueira de Azevedo Jr http://taba.bio.br/

Exploring the Scoring Function azevedolab.net

azevedolab.net

Amaral MEA,, Nery LR, Leite CE, de Azevedo Junior WF, Campos MM. Pre-clinical effects of metformin and aspirin on the cell lines of different breast cancer subtypes. Invest New Drugs. 2018. doi: 10.1007/s10637-018-0568-y. PubMed PDF Levin NMB, Pintro VO, Bitencourt-Ferreira G, Mattos BB, Silvério AC, de Azevedo Jr. WF. Development of CDK-targeted scoring functions for prediction of binding affinity. Biophys Chem. 2018; 235: 1–8. https://doi.org/10.1016/j.bpc.2018.01.004 Link PDF Pintro VO, Azevedo WF. Optimized Virtual Screening Workflow. Towards Target-Based Polynomial Scoring Functions for HIV-1 Protease. Comb Chem High Throughput Screen. 2017; 20(9): 820-827. PubMed PDF de Ávila MB, Xavier MM, Pintro VO, de Azevedo WF. Supervised machine learning techniques to predict binding affinity. A study for cyclin-dependent kinase 2. Biochem Biophys Res Commun. 2017; 494: 305-310. PubMed PDF

Usos Recentes do SAnDReS azevedolab.net

azevedolab.net


https://www.researchgate.net/publication/322888907_Pre-clinical_effects_of_metformin_and_aspirin_on_the_cell_lines_of_different_breast_cancer_subtypes






https://www.researchgate.net/profile/Walter_De_Azevedo_Jr2/publication/321220127_Optimized_Virtual_Screening_Workflow_Towards_Target-Based_Polynomial_Scoring_Functions_for_HIV-1_Protease/links/5a35131b0f7e9b10d845055a/Optimized-Virtual-Screening-Workflow-Towards-Target-Based-Polynomial-Scoring-Functions-for-HIV-1-Protease.pdf


https://www.researchgate.net/publication/320354236_Supervised_machine_learning_techniques_to_predict_binding_affinity_A_study_for_cyclin-dependent_kinase_2

Freitas PG, Elias TC, Pinto IA, Costa LT, de Carvalho PVSD, Omote DQ, Camps I, Ishikawa T, Arcuri HA, Vinga S, Oliveira AL, Junior WFA, da Silveira NJF. Computational Approach to the Discovery of Phytochemical Molecules with Therapeutic Potential Targets to the PKCZ protein. Letters in Drug Design & Discovery 2017 DOI: 10.2174/1570180814666170810120150 Link to the Paper Heck GS, Pintro VO, Pereira RR, de Ávila MB, Levin NMB, de Azevedo WF. Supervised Machine Learning Methods Applied to Predict Ligand-Binding Affinity. Curr Med Chem. 2017; 24(23): 2459-2470. PubMed PDF Levin NM, Pintro VO, de Ávila MB, de Mattos BB, De Azevedo WF Jr. Understanding the Structural Basis for Inhibition of Cyclin-Dependent Kinases. New Pieces in the Molecular Puzzle. Curr Drug Targets. 2017; 18(9): 1104-

1111. PubMed PDF ...

Usos Recentes do SAnDReS azevedolab.net

azevedolab.net

http://www.eurekaselect.com/154827







https://www.researchgate.net/publication/317848912_Supervised_Machine_Learning_Methods_Applied_to_Predict_Ligand-Binding_Affinity


https://www.researchgate.net/publication/310458926_Understanding_the_Structural_Basis_for_Inhibition_of_Cyclin-Dependent_Kinases_New_Pieces_in_the_Molecular_Puzzle

Considerações Finais

SAnDReS é um ambiente integrado para análise de resultados de docking

molecular e desenvolvimento de modelos de aprendizado de máquina;

SAnDReS é resultado de 3 anos de pesquisa e desenvolvimento;

SAnDReS apresenta mais de 25 mil linhas de código;

SAnDReS consumiu mais 12 mil horas/homem de trabalho;

SAnDReS é capaz de gerar modelos de aprendizado de máquina direcionados ao

sistema biológico de interesse;

Estudo da CDKcom informação de IC50 foi capaz de propor novas funções escores

com melhor poder de previsão;

Aplicável para análise de resultados gerados por qualquer programa de docking

molecular;

Disponível em: sandres.net

azevedolab.net

azevedolab.net

azevedolab.net

azevedolab.net

Nayara, Gabriela Heck, Carminha, Walter, Maurício e Val

Gabriela Bitencourt e Bruna

azevedolab.net

Prof. Dr. Ivan Cunha Bustamante Filho

Univates

CNPq ( Processo número: 308883/2014-4)

azevedolab.net

Dr. Walter Filgueira de Azevedo Jr.

Doutor em Ciências – Física Aplicada – Universidade de São Paulo –USP

Pesquisador Visitante na Universidade da Califórnia em Berkeley-EUA

Livre-Docente em Física – Universidade Estatual Paulista – UNESP

Editor Regional da Revista Current Drug Targets

Editor de Seção (Bioinformatics Applied to Drug Design) da Current

Mediicinal Chemistry

Membro do Corpo Editorial da Revista Current Bioinformatics

Pesquisador nível 1B do CNPq

E-mail: [email protected]

Facebook: https://facebook.com/Prof.Walter

Site: http://azevedolab.net

mailto:[email protected]













https://facebook.com/Prof.Walter












http://azevedolab.net/













Documents

azevedolab.net Biologia de Sistemas … › resources › Univates_2018_03_22...• Apresentação • Biologia de Sistemas Computacional • Docking Molecular • CDK • Projeto