Introduction to Virtual Screening

You may have noticed that technical words like “Virtual Screening” or “Target fishing” are all over this website, and you may not really know what they mean. This page is here to help you, and explain some of these concepts.

To start, we might have la look into what a “Virtual Screening” is. For a definition, we could have this one:

Virtual Screening, or “VS” for short, is a computational technique mainly used in the search for bioactive compounds in drug discovery, agrochemistry, phytochemistry, nutraceutricals, cosmetics. It consists in searching in libraries of small molecules in order to identify those structures which are most likely to bind to a usually larger target, typically a protein receptor or enzyme, nucleic acids or some other more complex macromolecules.

Translating this into plain language, it means that we use computers to calculate how well some small chemical molecule fits against, or interacts synergically with other big molecules. Maybe this does not sound like a big thing, but, there is definitely more to it than it seems. It means you can do all of the following (try to find your case!):

  1. You can try to fit a small molecule at a specific spot of a big molecule’s surface, and “measure” how well it binds to the surface on that spot. This is what we call a “docking“, and it is the most basic calculation we do. It will usually give us information of which parts of the small molecule bind to which parts of the big one.

    Image rendered from the results of a “Blind Docking” calculation.

  2. Then, you can take a big molecule, and a small molecule, and you can try to find where on the surface of the big molecule the smallest molecule binds better. This is what we call a “blind docking“, and takes a while longer than a simple “docking” to calculate.
  3. Also, you can take a (large) library of small molecules, and you can try to figure out which of them binds better to your big molecule. Or you can even rank them! The only things you need are the big molecule, and a collection of small molecules of your choice. Fortunately, there are large collections of such small molecules available freely, such as ZINC, pubchem or chembl databases. Or we can also design a custom library for you, of course. And then, you need some time at a supercomputer to make all these calculations: the more molecules you try, the more time you need. This kind of essay is what we properly call a “Virtual Screening“.
  4. As you can imagine, we can mix up what we describe in 2. and 3., and that takes LOTS of computer time. This would be a “blind docking virtual screening“.
  5. It also works in the other way: we can have a small molecule, and try it onto several different big molecules, to see if it successfully binds to any of them. This is called “Target fishing“.

Of course, as we are talking about computers, every time we say that we need “a molecule”, we mean that we need “a computer model of a molecule“. This means that we have to build a representation of each of these molecules (most of the times, such representations already existe, and we only have to find them). But this is not everything we can do: these computer models, have to capture the properties of the molecules. We are also able to compare the small molecules we were talking about by their properties.

So, if we know that a molecule is able to do something (like binding to a big molecule) because it has certain properties, we could also think that other molecules with similar properties (structural, electrostatic, pharmacophoric, etc) could do the same. Even if the molecule looks completely different. And we can do that: we can also compare sets of small molecules based on properties as shape, electrostatic charges, polarity, etc. This is what we call “search of non-structural analogues

And finally, we can “observe” what happens to a molecule over time when we put it in a certain situation. We can do this by making “molecular dynamics” simulations, and these may help us understand processes that happen over time and to elucidate biological mechanisms of action.


Ok. This is what we can do. But, why would we want to do any of that?

It is easy: Figure our “big molecule” is a protein responsible for some disease, and by binding a small molecule you are able to deactivate it, and cure your disease. You have just found a lead on a new drug.

Or, maybe your “big molecule” is an enzyme that is vital for some fungus or insect, and by binding the small molecule you are able to prevent it from growing. Now you have a fungicide, or an insecticide!

Also, it works not only for proteins or enzymes: we can also calculate binding affinities to other kind of molecules, i.e. cyclodextrins, so that we can make calculations for encapsulating smaller compounds, so that, i.e. they turn to be soluble in water.

These are only very basic things we can do, and just some very simple examples. If you are in doubt, or still do not get an idea of how all of this works, Contact us! We will be happy to help you out, and it will also help us to improve this page!

Was this information useful?
Please help us improve!

Take our quick survey
Was this information useful?


Please click this text to help us improve
by taking our quick survey.