Virtual Screening
Structure Based Virtual Screening
- Careful protein preparation – Know your target
- Careful ligand preparation – Enumerate states and
conformations - Pilot screening – Know the best combination of constraints
and scores - Screening
- Post-screening processing
- Purchase and assay
###Questions we will be Asking Today
• Do we have a crystal structure or an homology model?
• Is the target drug-able?
• What is the quality of the model - electron density?
• Understanding the binding site: big small buried
pockets general-properties?
• Are there known ligands for this target, and how can I use that information?
• How to screen compounds
• How to improve the quality of screened outputs?
• How to filter and cluster the output from screen to
manageable numbers for synthesis or purchase
There are numerous fundamental issues that should be examined when considering a biological target for SBVS; for example, the druggability of the receptor, the choice of binding site, the selection of the most relevant protein structure, incorporating receptor flexibility, suitable assignment of protonation states, and consideration of water molecules in a binding site, to name a few.
Generic framework for in silico small-molecule screening
Glaab, E. (2015). Building a virtual ligand screening pipeline using free software: a survey. Briefings in Bioinformatics, 17(2), 352-366.
Schrödinger
Workflow
- ligand preparation LigPrep,
- filtering using propfilter on QikProp properties or other structural
properties - Glide docking at the three accuracy levels:
HTVS 3-5 secs/lig
SP 30-50secs/lig
XP 3-5mins/lig
####1. Protein preparation wizard – prepare and repair PDB structures
• Cleaning up raw PDB files
– Assign bond order
– Protonation(Add hydrogen atoms)
– Delete unwanted part of the system(counterions,artifacts of crystallography,water)
– Optimize the hydrogen bond networks (flip of residues like ASN, GLN, tautomer
determination: HIE, HID or protonation state HIP…)
– Remove putative clashes in your structure (ideally with diffraction data)
• Missing information
– Important side-chains are missing
– Important loops are missing
QikProp is a quick, accurate, easy-to-use absorption, distribution, metabolism, and excretion
(ADME) prediction program
2.
two-dimensional similarity methods and shape or drug-like filters to
reduce the number of database compounds for the
time-consuming steps of flexible docking
The Virtual Screening Workflow offers three choices for prefiltering ligands:
Lipinski’s Rule of 5
Removing ligands with reactive functional groups
The functional groups that are considered reactive are:
• Acyl halides
• Sulfonyl halides
• Sulfinyl halides
• Sulfenyl halides
• Alkyl halides without fluorine
• Anhydrides
• Perhalomethylketones
• Aldehydes
• Formates
• Peroxides
• R-S-O-R
• Isothiocyanates
• Isocyanates
• Phosphinyl halides
• Phosphonyl halides
• Alkali metals
• Alkaline-earth metals
• Lanthanide series metals
• Actinide series metals
• Transition metals
• Other metals
• Toxic nonmetals
• Noble gases
• Carbodiimides
• Silyl enol ethers
• Nitroalkanes
• Phosphines
• Alkyl sulfonates
• Epoxides
• Azides
• Diazonium compounds
• Isonitriles
• Halopyrimidines
• 1,2-Dicarbonyls
• Michael acceptors
• Beta-heterosubstituted carbonyls
• Diazo compounds
• R-N-S-R
• Disulfides
Prefiltering with a custom filter
Set-up:
Compound Library:
Video:ZINC-How to download a database for docking
ADME–>Lipinski’s Rule of 5 –> PAINS
QikProp
ADME:
*absorption distribution metabolism excretion
Nearly 40% of drug candidates fail in clinical trials due to poor ADME (absorption, distribution, metabolism, and excretion) properties. These late-stage failures contribute significantly to the rapidly escalating cost of new drug development. The ability to detect problematic candidates early can dramatically reduce the amount of wasted time and resources, and streamline the overall development process.
Accurate prediction of ADME properties prior to expensive experimental procedures, such as HTS, can eliminate unnecessary testing on compounds that will ultimately fail; ADME prediction can also be used to focus lead optimization efforts to enhance the desired properties of a given compound. Finally, incorporating ADME predictions as a part of the development process can generate lead compounds that are more likely to exhibit satisfactory ADME performances during clinical trials.
Lipinski’s rule of five.
The rules are: mol_MW < 500, QPlogPo/w < 5, donorHB ≤ 5,
accptHB ≤ 10. Compounds that satisfy these rules are considered druglike. (The “five” refers to the limits, which are multiples of 5.)
PAINS (pan assay interference compounds )
The PAINS filter is available from the Structure menu. After opening Canvas and importing the compounds into the Canvas project, choose Structure Filter → Use → Pains1, Pains2, or Pains3. See the Canvas User Manual for more information.
Other filter tool:https://mobyle.rpbs.univ-paris-diderot.fr/cgi-bin/portal.py#forms::FAF-Drugs4
REOS（Rapid Elimination of Swill）
REOS (rapid elimination of swill) filter is based on more than 100 SMARTS strings collected from literature data describing non-druglike functionalities associated with promiscuous ligands and frequent hitters. Efficiency of lead optimization efforts can be significantly increased by eliminating these problematic compounds to focus on top quality drug candidates only.
When to use
If you are looking for drug candidates or even just suitable probe molecules to analyze a biological target, it is highly recommended to exclude molecules containing reactive functional groups as these can be frequently associated with toxicity and lack of selectivity. Conclusions drawn from biological experiments with compounds containing such functional groups might therefore be misleading. Unless you are looking for reagents for further synthesis or for building a combinatorial library, the REOS filter is highly recommended to filter out problematic compounds.
e-Pharmacophores
Ligand-based pharmacophore modeling and structure-based protein-ligand docking are both recognized as integral parts of drug discovery, each method offering particular strengths. Ligand-based technologies, such as 3D-pharmacophore modeling, are fast and thus useful for quickly screening large compound databases. On the other hand, structure-based approaches can yield more diverse actives and lead to important target insights, but can be time-consuming. The e-Pharmacophores method achieves the advantages of both ligand- and structure-based approaches by generating energetically optimized, structure-based pharmacophores that can be used to rapidly screen millions of compounds.
Human Metabolite Database (HMDB) [26]: This
library with 2,462 compounds contains information
about small molecule metabolites found in the human
body
protein:
Alpha-ketoglutarate-dependent dioxygenase FTO
is a protein implicated in the development of obesity.
FTO is the strongest genetic predictor of increased
body weight.
3LFM
32 * 32 * 32 =32768 $A^3$, centered
around the enzyme’s active site
The blood-brain barrier (BBB) filter
References and sources:
Structure-Based Virtual Screening Using Glide
Virtual Screening with Glide-byThijs Beuming
Building a virtual ligand screening pipeline using free software: a survey
Schrödinger药物虚拟筛选流程模块在大学生物和化学信息学教学中的应用
https://doc.mcule.com/doku.php?id=reos
The three-dimensional structure of NadE was submitted in DockBlaster (http://blaster.docking.org/) for performing virtual screening against ZINC Database.
