rpbs logoFAF-Drugs: Free ADME/tox Filtering

In silico screening has become an essential tool in the drug discovery process. Compound collections in 3D are needed to carry out such in silico experiments. Yet, it has been recognized that absorption, distribution, metabolism, excretion and toxicity (ADME/tox) are key properties that need to be considered early on, even during the database preparation stage and often tailored to a given project. FAFDrugs is an online service that allows users to process their own compound collections via simple ADME/Tox filtering rules such as molecular weight, polar surface area, logP or number of rotatable bonds.

Users can filter their own compound collections and use additional tools and compound collections for virtual ligand screening projects.
Salt removal
remove salt from
compound description
(see below)
ADME/tox Filter
filter online your own collections of compounds.

 Collections
Collections of compounds prefiltered at different levels.
 More information


Pocket finder: we have implemented PASS, a method to identify protein cavities
Ref. Fast Prediction and Visualization of Protein Binding Pockets With PASS
G. Patrick Brady, Jr. and Pieter F.W. Stouten Journal of Computer-Aided Molecular Design, 14: 383-401, 2000

Online XlogP calculations (computed with XScore, sw16.im.med.umich.edu/software/xtool). A comparison of experimental versus calculated values for 109 compounds is shown here

Salt removal: a tool to remove salts sometimes associated in compound description. Collections of compounds should be desalted before ADMETox filtering, otherwise some bias will be introduced for molecular weight calculation, logP calculation, etc.

Online OpenBabel for file format conversion (openbabel.sourceforge.net)

Implementation of the Java Molecular Editor from Dr. Peter Ertl (Novartis Pharma AG, Basel, Switzerland) to draw molecules and obtain the
corresponding SMILES string.

Test sets for virtual ligand screening
These test sets include 6 different proteins (different folds and different nature of the binding pocket in term of open,
close, polar, hydrophobe) : ER, factor Xa, HIVprotease, Neuraminidase, RNase and TK, all in PDB format with the PDB ID
and corresponding ligands (active compound co-crystalized with these protein targets), about 10 ligands for each protein
in Mol2 format. These data can help to assess docking and scoring functions for a VLS project. These small actives can for
example be merged with larger banks distributed at RPBS and evaluation of docking and ranking performance can be carried out.
 
More information:


ADME/tox filtering Input:

Molecules in SMILES or 2D/3D SDF format are required.
Our service is built on Frowns (developed by Brian Kelley), a chemoinformatics toolkit.
The calculated molecular properties are:
a) Molecular weight
b) Hydrogen bond donors and acceptors
The number of hydrogen bond acceptors (sum of N + O) and hydrogen bond donors (sum of OH + NH)
c) Number of rigid bonds
d) Number of rings
e) Size of the rings
f) Number of rotatable bond
This is defined as any single non-ring bond, bounded to non-terminal heavy atom (i.e., non-hydrogen)(Veber, 2002)
The amide C-N bonds are not considered because of their high rotational energy barrier
g) Number of carbon atoms, Number of heteroatoms and ratio
h) Number of atom with a net charge
i) Sum of formal charges
j) The 2D polar surface area (Topological Polar Surface Area or TPSA)
The method described in (Ertl, 2000) has been implemented
k) Computation of XlogP (P = calculated octanol/water partition coefficient)
We use the XScore package (sw16.im.med.umich.edu/software/xtool) to compute XlogP as described in (Wang, 2000)
A comparison of experimental versus calculated values for 109 compounds is shown here
l) Atom filter
Molecules with some specific atoms can be filtered-out
(for instance molecules containing H, C, N, O, F, S, P, Cl, Br, I atoms are kept when using default parameters)

Ref.
Veber, D.F., Johnson, S.R., Cheng, H.Y., Smith, B.R., Ward, K.W. and Kopple, K.D. (2002) J Med Chem, 45, 2615-2623
Ertl, P., Rohde, B. and Selzer, P. (2000) J Med Chem, 43, 3714-3717
Wang, R., Gao, Y. and Lai, L. (2000) Perspectives in Drug Discovery and Design, 19, 47-66


Warnings:
The SDF format should be:

sdf format

Output:
2 files are generated, one called toxic.smi which contains molecules that do not pass the filters
and the file non_toxic.smi which contains molecules that do pass the filters, both are in CANSMILES
In addition, files with all computed properties can be obtained


 Collections

Compound collections
2D SDF files from http://www.ambinter.com/ (525000 molecules before filtering), http://www.asinex.com/ (365508 molecules before filtering),
http://chembridge.com/chembridge/ (429623 molecules before filtering and different subsets), http://dtp.nci.nih.gov/ (250251 molecules before filtering)
and http://www.specs.net/ (198462 molecules before filtering) were downloaded and converted with OpenBabel to SMILES.
Through a special agreement with OpenEye Scientific Software (http://www.eyesopen.com) we were allowed to filter these libraries with Filter (version 1.0.2),
a program to desalt the molecules and remove undesirable compounds based on LipinskiÕs RO5 (and related) and about 100 rules to eliminate compounds
with unstable/reactive chemical groups. Three different levels of filtering were used, very light, light and heavy (see below and Filter user guide).
Then, the OpenEyeÕs Omega program was used to generate 3D models, either single conformation or up to 50 conformations for each molecule (multiconformer files)
that passed the filters. Hydrogen atoms and Gasteiger partial charges were added by Omega during the process.

The files usually contain 15000 molecules and are saved in Mol2 format (see information about the format at http://www.tripos.com),
the filtered molecules in SMILES format are also available. For the time being, we did not generate multiple protonation states
nor considered enumerating different isomeric forms. The chemical supplier ID numbers are present in the Mol2 files and Smiles files whenever available.

Example of filter_light input

This file defines the rules for filtering multi-structure files based on
properties and substructure patterns.
NOTE: this rule file is suitable for drug-like filtering and may be
too restrictive for reagent filtering.
These rules include the contributions of countless chemists & modellers including:
Rule of 5:
Lipinski, C, et al., Adv.Drug Deliv. Rev., 23:3, 1997.
General Filtering:
Oprea, T., JCAMD 14:251-264, 2000.
XLogP:
Wang, R, Ying, Fu, & Lai, Luhua, J.Chem. Inf. Comput. Sci., 37:615-621, 1997.
Polar Surface Area
Peter Ertl, Bernhard Rohde, & Paul Slzer, J. Med. Chem. 43:3714-3717, 2000.
David E. Clark, Journal of Pharmaceutical Sciences, 88(8):807-814, 1999.
MIN_MOLWT 200 Minimum molecular weight
MAX_MOLWT 600 Maximum molecular weight
MIN_SSSR_RINGS 0 Minumum number of SSSR rings
MAX_SSSR_RINGS 7 Maximum number of SSSR rings
MAX_RING_SIZE 20 Maximum size of any SSSR ring

MIN_CARBONS 5 Minimum number of carbons
MIN_HETEROATOMS 2 Minimum number of heteroatoms
MIN_Het_C_Ratio 0.10 Minimum heteroatom to carbon ratio
MAX_Het_C_Ratio 1.0 Maximum heteroatom to carbon ratio

ADJUST_ROT_FOR_RING_TRUE BOOLEAN for weather to estimate degrees of freedom in rings

MIN_ROT_BONDS 0 Minimum number of rotatable bonds
MAX_ROT_BONDS 20 Maximum number of rotatable bonds

MIN_RIGID_BONDS 0 Minimum number of rigid bonds
MAX_RIGID_BONDS 50 Maximum number of rigid bonds

MIN_HBOND_DONORS 0 Minimum number of hydrogen-bond donors
MAX_HBOND_DONORS 8 Minimum number of hydrogen-bond donors

MIN_HBOND_ACCEPTORS 0 Minimum number of hydrogen-bond acceptors
MAX_HBOND_ACCEPTORS 12 Minimum number of hydrogen-bond acceptors

MIN_COUNT_FORMAL_CRG 0 Minimum number formal charges
MAX_COUNT_FORMAL_CRG 3 Maximum number of formal charges
MIN_SUM_FORMAL_CRG 0 Minimum sum of formal charges
MAX_SUM_FORMAL_CRG 2 Maximum sum of formal charges

MIN_XLOGP -5.0 Minimum XLogP
MAX_XLOGP 6.0 Maximum XLogP

MIN_2D_PSA 0.0 Minimum 2-Dimensional (SMILES) Polar Surface Area
MAX_2D_PSA 160.0 Maximum 2-Dimensional (SMILES) Polar Surface Area
ALLOWED_ELEMENTS H,C,N,O,F,S,P,Cl,Br,I
ELIMINATE_METALS Sc,Ti,V,Cr,Mn,Fe,Co,Ni,Cu,Zn,Y,Zr,Nb,Mo,Tc,Ru,Rh,Pd,Ag,Cd
acceptable molecules must have less or equal instances of each of the patterns below
specific, undesirable functional groups
RULE 0 Carbazides
RULE 0 Acid_anhydrides
RULE 0 Pentafluorophenyl_esters
RULE 0 Paranitrophenyl_esters
RULE 0 HOBT_esters
RULE 0 Triflates
RULE 0 Lawesson_s_reagent
RULE 0 Phosphoramides
RULE 0 Aromatic_azides
RULE 0 Beta_carbonyl_quart_nitrogen
RULE 0 Acylhydrazide
RULE 0 Quarternary_C_Cl_I_P_or_S
RULE 0 Phosphoranes
RULE 0 Chloramidines
RULE 0 Nitroso
RULE 0 P_S_Halides
RULE 0 Carbodiimide
RULE 0 Isonitrile
RULE 0 Triacyloxime
RULE 0 Cyanohydrins
RULE 0 Acyl_cyanides
RULE 0 Sulfonyl_cyanides
RULE 0 Cyanophosphonates
RULE 0 Azocyanamides
RULE 0 Azoalkanals
RULE 0 Polyenes
RULE 0 Saponin_derivatives
RULE 0 Cytochalasin_derivatives
RULE 0 Cycloheximide_derivatives
RULE 0 Monensin_derivatives
RULE 0 Cyanidin_derivatives
RULE 0 Squalestatin_derivatives
functional groups which often eliminate compounds from consideration
RULE 0 acid_halide
RULE 0 aldehyde
RULE 0 alkyl_halide
RULE 0 anhydride
RULE 0 azide
RULE 0 azo
RULE 0 di_peptide
RULE 0 long_aliphatic_chain //(>7 atoms)
RULE 0 michael_acceptor
RULE 0 beta_halo_carbonyl
RULE 0 nitro
RULE 0 peroxide
RULE 0 phosphonic_acid
RULE 0 phosphonic_ester
RULE 0 phosphoric_acid
RULE 0 phosphoric_ester
RULE 0 sulfonic_acid
RULE 0 sulfonic_ester
RULE 0 triphenyl_phosphene
RULE 0 unbranched_chain //(>4_atoms)
RULE 0 epoxide
RULE 0 hetero_hetero
RULE 0 sulfonyl_halide
RULE 0 halopyrimidine
RULE 0 perhalo_ketone
RULE 0 methyl_ketone
RULE 0 aziridine
RULE 0 imine
RULE 0 oxalyl

the dye group includes a set of patterns which describe all cpds with colors in their names from the ACD98.2

RULE 0 dye
functional groups which are allowed, but may not be wanted in high quantities
common functional groups

RULE 6 alcohol
RULE 8 alkene
RULE 4 amide
RULE 4 amino_acid
RULE 4 amine
RULE 4 primary_amine
RULE 4 secondary_amine
RULE 4 tertiary_amine
RULE 4 carboxylic_acid
RULE 6 halide
RULE 1 iodine
RULE 4 ketone
RULE 4 phenol
other functional groups
RULE 4 alkyne
RULE 4 aniline
RULE 4 aryl_halide
RULE 4 carbamate
RULE 4 ester
RULE 4 ether
RULE 4 hydrazine
RULE 4 hydrazone
RULE 4 hydroxylamine
RULE 4 nitrile
RULE 4 sulfide
RULE 4 sulfone
RULE 4 sulfoxide
RULE 4 thiourea
RULE 4 thioamide
RULE 4 thiol
RULE 4 urea