I have a set of 30 chemically similar compounds, i want to do virtual screening,
should i start docking studies after drawing their structure or should perform other any other analysis first?
What that analysis can be ?
Dear Ayzeha,
I did the same kind of work in my masters. I did first virtual screening based on pharmacophore based approach. Then I design some compounds using the final hits. The designed compounds have similar scaffold to that of the parent compound only the substitution is different. Then I performed simply molecular docking studies to see whether these compounds are binds in a correct mode and provides the same interaction as that of parent compound. If the scaffold is same in your case and only there is a difference in substitution then I will suggest you to do molecular docking and compare the result. I guess you can simply distinguish them on the basis of docking score, interactions and binding pocket of active site. You can simply discard those which are not fitting to the binding pocket by visualizing the docking complex.
Best of luck!!
Gaurao
Dear Ayzeha,
I did the same kind of work in my masters. I did first virtual screening based on pharmacophore based approach. Then I design some compounds using the final hits. The designed compounds have similar scaffold to that of the parent compound only the substitution is different. Then I performed simply molecular docking studies to see whether these compounds are binds in a correct mode and provides the same interaction as that of parent compound. If the scaffold is same in your case and only there is a difference in substitution then I will suggest you to do molecular docking and compare the result. I guess you can simply distinguish them on the basis of docking score, interactions and binding pocket of active site. You can simply discard those which are not fitting to the binding pocket by visualizing the docking complex.
Best of luck!!
Gaurao
It is a good idea to consider your input structures prior to docking. First see if there are any crystal or NMR structures available for your compounds. If there are and your compounds are simple analogues thereof, then you have a template to help guide your 3D structure building of the remaining compounds. If an experimentally determined 3D structure does not exist, then you could do several things. First consider basic conformational analysis as one learns in an organic course (e.g. cis versus trans stability, anomeric effect). Depending on the resources you have you could conduct some type of quantum mechanical or semi-empirical study to identify energetically low lying minima, and use these as your input into the docking program. Some “off-the-shelf” programs will also allow you to do a molecular mechanics Monte Carlo search for possible minima. Two more things to consider are if the molecules have enantiomers, and if they could exist in a (de-)protonated state. I believe the Schrodinger software suite has some of these capabilities.
Thanku so much. i have drawn the 3d structures of all ligands, Can you please tell which free or trail software could i use for screening these ligands based on pharmacophore based approach?
Dear Ayzeha,
I did virtual screening using Discovery studio2.5 and using LigandScout3.0, both are commercial software. I don't have much detail idea about freely available tools but I found few.
Please go through the links I hope some of them might be useful for you.
http://bioinfo3d.cs.tau.ac.il/pharma/about.html
http://anchorquery.ccbb.pitt.edu/
http://smoothdock.ccbb.pitt.edu/pharmer/
In addition I found one literature where they implemented PharmaGist online web based server for the development of pharmacophore and used this pharmacophore model as a query to screen Zinc database using Paramer. Instead of zinc database you can screen your own dataset.
Please find the article in the attachment.
If you are interested in rigid protein - flexible ligand docking methods, the chains that are not involving any structural interaction with active site region can be deleted.
I do hope you found the protein of interest to carry out virtual screening (here, its actually docking of 30 similar compounds). If not you can try reverse-docking approach or STITCH (interaction networks of chemicals and proteins) to find the protein of interest.
Its good to energy minimize the small molecules before docking (or you can proceed without that)
Once you identify the best molecule (docking score, based on free energy of binding, clustering of docking studies and interactions), u can find similar molecules to carry out secondary virtual screening on the small molecule databases like ZINC database or database of your interest.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology