Validated targets annotated space

1.

VALIDATED TARGETS
ANNOTATED SPACE

2.

conception
Train set:
1) A set of 900 drugs launched or in clinical trials over the past 10 years has been assembled (38 targets)
2) For each structure 300 3D-conformations were generated
3) 100 the most divers conformations were selected for each sample
4) CATS 3D descriptors were calculated for all conformations
Target set:
1) Soft MCFs and Drug-likeness filters were applied to compounds available in the ChemDiv collection (>1.5 M cmpds)
2) For each molecule 3D-conformation was generated in Corina Software - 256к.
3) 100 conformations were generated for the starting conformation, and about 10 maximally different ones were selected
(of total - 2.5 M samples)
4) CATS 3D descriptors were calculated for all conformations
Algorithm:
1) A neural network was trained for the prediction of the target activity. Activity prediction (0 or 1) is performed by the model
independently for each target
2) Using iterative optimization, individual binarization thresholds were selected to make the final decision for each target
3) Numerical metrics were calculated on a validation set of 10% of the training set. In addition, the quality of the model was
assessed on 10 compounds from an independent sample.

3.

Strategy
molecules
launched or in
clinical trials
Deep AI
Train
Target set
ChemDiv collection
Model
library
18K compounds

4.

MoA predictor
Targets
5-HT4 Receptor
5-HT7 Receptor
Bromodomain-Containing Protein 4 (BRD4; HUNK1)
Bruton's Tyrosine Kinase (BTK)
Calcium Channel Blockers
Cyclin-Dependent Kinase 4 (CDK4)
Cyclin-Dependent Kinase 6 (CDK6)
Dipeptidyl Peptidase IV (CD26; DPP-IV; DP-IV)
Dopamine D2 Receptor (DRD2)
E3 Ubiquitin-Protein Ligase Mdm2 (Hdm2)
Farnesoid X Receptor (FXR)
Free Fatty Acid Receptor 1 (FFAR1; GPR40)
GTPase KRAS
HER2 (erbB2)
precision
1
1
recall
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.7
0.75
0.666667
0.666667
1
1
1
0.666667
1
1
1
Targets
precision
recall
CDK4/Cyclin D1
0.666667
1
CDK9/Cyclin T1
Cystic Fibrosis Transmembrane Conductance
Regulator (CFTR)
0.666667
1
0.6
1
0.75
0.857143
EGFR
1
1
Estrogen Receptor (ER)
0.5
1
Fibroblast Growth Factor Receptor 1 (FGFR1)
1
0.5
Fibroblast Growth Factor Receptor 2 (FGFR2)
1
0.5
Fibroblast Growth Factor Receptor 4 (FGFR4)
0.5
1
FLT3 (FLK2/STK1)
0.5
0.285714
Hepatocyte Growth Factor Receptor (HGFR; MET)
0
0
Phosphatidylinositol 3-Kinase (PI3K)
1
0.75
Poly(ADP-ribose) Polymerase (PARP)
1
1
Tyrosine-Protein Kinase JAK 1, 2
Vascular Endothelial Growth Factor Receptor 2
(VEGFR2; FLK1)
1
1
1
0.4
Tyrosine-Protein Kinase JAK3
1
1
Dopamine D3 Receptor (DRD3)
Mitogen-Activated Protein Kinase 3 (MAPK1; ERK2)
mu-Opioid Receptor
Muscarinic Receptor
1
1
1
1
1
1
Non-Structural Protein 5A (NS5A)
Orexin OX-2 Receptor
Selective Estrogen Receptor Degradation Inducers
(SERD)
1
1
1
1
Tyrosine-Protein Kinase Receptor UFO (AXL)
Sodium Channel Protein Type 9 Subunit alpha
(SCN9A; Nav1.7)
1
1
1
0.666667
1
1
Protein Cereblon (CRBN) Modulators
1
1

5.

Statistics
NN
Setup
Precision
Recall
V3A
Avg prob > indiv val
0.81
0.71

6.

Statistics
Phencynonate
hydrochloride
M-201
Almonertinib
mesilate
LB-102
ARQ-252
Zorifertinib
Seltorexant
mol
Phencynonate
hydrochloride
LB-102
M-201
LY-2922083
Almonertinib
mesilate
ARQ-252
Seltorexant
Zorifertinib
LY-2922083
true_targets
predicted_targets
['Muscarinic Receptor Antagonists']
[('Apoptosis Inducers',), ('Muscarinic Receptor Antagonists',)]
['Dopamine D2 Receptor (DRD2) Antagonists', 'Dopamine D3 Receptor (DRD3) [('Dopamine D2 Receptor (DRD2) Antagonists',), ('Dopamine D3 Receptor
Antagonists', '5-HT7 Receptor Antagonists']
(DRD3) Antagonists',), ('5-HT7 Receptor Antagonists',)]
['Calcium Channel Blockers']
[('Calcium Channel Blockers',)]
['Free Fatty Acid Receptor 1 (FFAR1; GPR40) Agonists', 'Glucose Lowering
Agents']
[('Free Fatty Acid Receptor 1 (FFAR1; GPR40) Agonists',), ('Glucose
Lowering Agents',), ('Cystic Fibrosis Transmembrane Conductance
Regulator (CFTR) (DeltaF508 Mutant) Correctors',)]
['EGFR (HER1; erbB1) Inhibitors', 'EGFR (Thr790Met Mutant) Inhibitors']
['Tyrosine-Protein Kinase JAK1 Inhibitors']
['Orexin OX-2 Receptor Antagonists']
['EGFR (HER1; erbB1) Inhibitors']
[('EGFR (Thr790Met Mutant) Inhibitors',)]
[('Tyrosine-Protein Kinase JAK1 Inhibitors',)]
[('Orexin OX-2 Receptor Antagonists',)]
[('EGFR (HER1; erbB1) Inhibitors',)]

7.

Representative Examples
2D DIVERSITY

8.

Thank
You!