Hi,
I had a shot at ANNalog by submitting losartan as query:
annalog-generate -i losartan.smi -n 50 -m beam -o losartan_50_beam.tsv --device cuda
Frankly I find the results unexpected to say the least (see below). The top-ranked generated molecule looks very odd IMO. I did a substructure search in ChEMBL on the 4-membered ring with 3 nitrogens (C1=NN=N1) and (as expected) get zero hits.
Please advise on how to tweak settings to get more reasonable results.
Thanks/Evert
input_smiles rank generated_smiles score check_scaffold check_skeleton check_ring_systems check_structural_alerts check_lacan
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 1 CCCCC1=NC(=C(N2CCCCC2)C2=CC=CC=C2C3=NN=N3)CO1 -7.370646662428044 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 2 CCCCC1=NC(=C(N2CC3=CC=CC=C3C3=NN=N3)C2)CO1 -7.438992510076787 False False False 9 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 3 CCCCC1=NC(=C(N2CCCCC2)C2=CC=CC=C2C3=NNN=N3)CO1 -7.621849828301492 False False False 0 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 4 C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)C -7.896411719884782 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 5 CCCCC1=NC(=C(N2CCCC2)C2=CC=CC=C2C3=NN=N3)CO1 -7.946875530909892 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 6 CCCCC1=NC(=C(N2CC=C(C=C2)C2=CC=CC=C2C3=NN=N3)CO)CO1 -8.084508885429386 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 7 C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)Cl -8.131075549092202 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 8 CCCCC1=NC(=C(N2CCCC2)C2=CC=CC=C2C3=NNN=N3)CO1 -8.18155053797092 False False False 0 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 9 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\O -8.332238370439882 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 10 C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)CO -8.393718364801316 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 11 C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\O -8.487318041887193 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 12 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)Cl -8.886250505111093 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 13 c1ccccc1CC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\O -8.893988891948538 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 14 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)C -9.065679872295732 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 15 CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C2=CC=CC=CC=C2C1=NN=N1)CO)Cl -9.078178473086155 False False False 4 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 16 c1ccccc1CC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)Cl -9.100911271063524 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 17 CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C2=CC=CC=C2C1=NN=N1)CO)Cl -9.196051581492839 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 18 C(CCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)CO)Cl -9.650958922924474 False False False 9 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 19 CC(C)C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)Cl -9.74157674414164 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 20 c1ccccc1CCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\O -9.794060989726859 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 21 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)\O -9.819780210269528 False True True 0 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 22 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl -9.896669653062418 False True True 0 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 23 c1ccccc1CCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)Cl -9.942183686172939 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 24 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)C -9.972666004909115 False True True 0 1.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 25 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)F -10.000733437771714 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 26 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C2=CC=CC=C2C1=NN=N1)CO)Cl -10.003846071367889 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 27 CC(CC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)CO)Cl -10.044025120256265 False False False 7 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 28 c1ccccc1CC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)C -10.077354177345114 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 29 CCCCC1=NC(=C(N2CC3=CC=C(C=C3)C3=CC=CC=C3C4=NN=N4)CO)COC12 -10.125140683136124 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 30 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C2=CC=CC=CC=C2C1=NN=N1)CO)Cl -10.178661473350076 False False False 4 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 31 c1cc(F)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\O -10.210662088461504 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 32 c1ccccc1CC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)\O -10.739072198284703 False True True 0 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 33 c1cc(Cl)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\O -10.8878274687886 False False False 1 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 34 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(=O)C -11.48895775444089 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 35 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(C)=O -11.822696716062637 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 36 c1cc(F)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(C)=O -12.615123438753471 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 37 c1cc(Cl)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(C)=O -12.724961310137587 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 38 c1cc(C)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(C)=O -12.790548102253524 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 39 c1cc(Cl)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(=O)C -12.854483520364738 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 40 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN(C)C(=O)N4)CO)\O -12.88633275917573 False False True 0 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 41 c1cc(F)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(=O)C -12.887201947280118 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 42 c1cc(OC)ccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(C)=O -13.306095810494298 False False False 2 0.0
CCCCC1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NNN=N4)CO)Cl 43 c1ccccc1C1=NC(=C(N1CC2=CC=C(C=C2)C3=CC=CC=C3C4=NN=N4)CO)\OC(C)=C(C)C -13.806722997511542 False False False 3 0.0
Hi,
I had a shot at ANNalog by submitting losartan as query:
annalog-generate -i losartan.smi -n 50 -m beam -o losartan_50_beam.tsv --device cudaFrankly I find the results unexpected to say the least (see below). The top-ranked generated molecule looks very odd IMO. I did a substructure search in ChEMBL on the 4-membered ring with 3 nitrogens (C1=NN=N1) and (as expected) get zero hits.
Please advise on how to tweak settings to get more reasonable results.
Thanks/Evert