Skip to main content

Natural Product-likeness in ChEMBL

By

Recently, we included a Natural Product-likeness score for chemical compounds stored in ChEMBL. We made use of an algorithm published by Peter Ertl, Silvio Roggo and Ansgar Schuffenhauer in 2008. 

Whereas the original version of this algorithm used a commercial data set of Natural Product molecules for training the algorithm (the CRC Dictionary of Natural Products) and an in-house library of synthetic molecules as a negative set, we used Greg Landrum's RDKit implementation which is based on ~50,000 natural products collected from various open databases and ~1 million drug-like molecules from ZINC as a "non-Natural Product background".

After including the new score into ChEMBL, we were interested to see whether the results look meaningful. We had a handful of simple questions:

  1. How is Natural Product-likeness distributed in ChEMBL and how does this compare to Natural Product-likeness for "real" NPs?
  2. Can we observe any difference in Natural Product-likeness for compounds from different MedChem Journals in ChEMBL?
  3. How do typical Natural Products stored in ChEMBL look like?

In general, Natural Product-likeness for compounds stored in ChEMBL 32 ranges from -4.1 to 4.1 with a median value of around -1:

Comparison to "real" Natural Products

How does this relate to Natural Product-likeness of "real" Natural Products? To answer this question we need a ground truth dataset of Natural Products. We retrieved this dataset from ChEBI, our dictionary and ontology of Chemical Entities of Biological Interest, by querying for molecules that are defined as "metabolite" (CHEBI:25212) while at the same time not being a "drug metabolite" (CHEBI:49103) on the web interface ("Advanced Search" functionality):

ChEBI "Advanced Search" panel with the query used to retrieve the dataset of Natural Products

This gave a dataset of around 22K unique ChEBI molecules which are Natural Products. One third of these molecules (7,601) are included in ChEMBL 32. Their Natural Product-likeness distribution looks like this, with a median Natural Product-likeness score of 1.47:

So how does that compare to the distributions for ChEMBL compounds? Looks like the algorithm captures Natural Products well and that the distribution we found in ChEMBL resembles the distributions that the developers of the original algorithm for calculating Natural Product-likeness found:

ChEMBL NP-likeness distribution reflects well the original findings by Ertl et al.; graph on the right reprinted with permission from the American Chemical Society, J Chem Inf Model 2008 48 (1), 68-74, DOI: 10.1021/ci700286. Copyright 2008 American Chemical Society.

Next, let's look at typical chemical comppounds from this set of > 7K Natural Product compounds in ChEMBL. The structures that are displayed are those with most entries from this subset in ChEMBL 32. It appears that not all of them do also possess an appropriate Natural Product-likeness score. Uric Acid, for example, seems to ressemble features of syntehtic molecules, therefore displaying a score of only -0.57.

Typical Natural Products from ChEBI/ChEMBL and their NP-likeness score

Natural Product-likeness in different Journals

Another interesting aspect to study is the variation of focus of chemical research that is published in different Journals. We can anticipate that chemical structures published in the Journal of Natural Products are showing a higher Natural Product-likeness on average than compounds being published in classical MedChem journals without a specific focus on Natural Products.

Our analysis suggests that this assumption is indeed correct, with chemicals from J Nat Prod possessing a median Natural Product-likeness score of .1.91 while for the other ChEMBL core journals this value ranges between -0.66 and -1.61:




Here are some typical representatives of molecules extracted from J Nat Prod which do possess a high Natural Product-likeness score (>3):

Typical Natural Products from J Nat Prod and their NP-likness score

What's next?

We hope the new implementation of a Natural Product-likeness score is useful to the drug discovery community and that it will spike some interesting science questions. Of course there are alternative algorithms available and there are some databases that specialised on Natural Products. We aim to also include a new Natural Product flag in the future, in addition to the score for Natural Product-likeness. It will be interesting to see how well the two approaches are matching up. However, the success of using a predictive algorithm for calculating a score as well as the flagging of "real" Natural Products in ChEMBL will always depend upon the ground truth dataset for Natural Products.

 

 

 

 

 

 

 

 

 

 

 

 

 


 



 













Comments

Post a Comment

Popular posts from this blog

Improvements in SureChEMBL's chemistry search and adoption of RDKit

By Eloy
    Dear SureChEMBL users, If you frequently rely on our "chemistry search" feature, today brings great news! We’ve recently implemented a major update that makes your search experience faster than ever. What's New? Last week, we upgraded our structure search engine by aligning it with the core code base used in ChEMBL . This update allows SureChEMBL to leverage our FPSim2 Python package , returning results in approximately one second. The similarity search relies on 256-bit RDKit -calculated ECFP4 fingerprints, and a single instance requires approximately 1 GB of RAM to run. SureChEMBL’s FPSim2 file is not currently available for download, but we are considering generating it periodicaly and have created it once for you to try in Google Colab ! For substructure searches, we now also use an RDKit -based solution via SubstructLibrary , which returns results several times faster than our previous implementation. Additionally, structure search results are now sorted by
Post a Comment
Read more

Improved querying for SureChEMBL

By
    Dear SureChEMBL users, Earlier this year we ran a survey to identify what you, the users, would like to see next in SureChEMBL. Thank you for offering your feedback! This gave us the opportunity to have some interesting discussions both internally and externally. While we can't publicly reveal precisely our plans for the coming months (everything will be delivered at the right time), we can at least say that improving the compound structure extraction quality is a priority. Unfortunately, the change won't happen overnight as reprocessing 167 millions patents takes a while. However, the good news is that the new generation of optical chemical structure recognition shows good performance, even for patent images! We hope we can share our results with you soon. So in the meantime, what are we doing? You may have noticed a few changes on the SureChEMBL main page. No more "Beta" flag since we consider the system to be stable enough (it does not mean that you will never
Post a Comment
Read more

ChEMBL brings drug bioactivity data to the Protein Data Bank in Europe

By
In the quest to develop new drugs, understanding the 3D structure of molecules is crucial. Resources like the Protein Data Bank in Europe (PDBe) and the Cambridge Structural Database (CSD) provide these 3D blueprints for many biological molecules. However, researchers also need to know how these molecules interact with their biological target – their bioactivity. ChEMBL is a treasure trove of bioactivity data for countless drug-like molecules. It tells us how strongly a molecule binds to a target, how it affects a biological process, and even how it might be metabolized. But here's the catch: while ChEMBL provides extensive information on a molecule's activity and cross references to other data sources, it doesn't always tell us if a 3D structure is available for a specific drug-target complex. This can be a roadblock for researchers who need that structural information to design effective drugs. Therefore, connecting ChEMBL data with resources like PDBe and CSD is essen
Post a Comment
Read more

ChEMBL 34 is out!

By
We are delighted to announce the release of ChEMBL 34, which includes a full update to drug and clinical candidate drug data. This version of the database, prepared on 28/03/2024 contains:         2,431,025 compounds (of which 2,409,270 have mol files)         3,106,257 compound records (non-unique compounds)         20,772,701 activities         1,644,390 assays         15,598 targets         89,892 documents Data can be downloaded from the ChEMBL FTP site:  https://ftp.ebi.ac.uk/pub/databases/chembl/ChEMBLdb/releases/chembl_34/ Please see ChEMBL_34 release notes for full details of all changes in this release:  https://ftp.ebi.ac.uk/pub/databases/chembl/ChEMBLdb/releases/chembl_34/chembl_34_release_notes.txt New Data Sources European Medicines Agency (src_id = 66): European Medicines Agency's data correspond to EMA drugs prior to 20 January 2023 (excluding vaccines). 71 out of the 882 newly added EMA drugs are only authorised by EMA, rather than from other regulatory bodies e.g.
Post a Comment
Read more

In search of the perfect assay description

By
Credit: Science biotech, CC BY-SA 4.0 Assays des cribe the experimental set-up when testing the activity of drug-like compounds against biological targets; they provide useful context for researchers interested in drug-target relationships. Ver sion 33 of ChEMBL contains 1.6 million diverse assays spanning ADMET, physicochemical, binding, functional and toxicity experiments. A set of well-defined and structured assay descriptions would be valuable for the drug discovery community, particularly for text mining and NLP projects. These would also support ChEMBL's ongoing efforts towards an  in vitro  assay classification. This Blog post will consider the features of the 'perfect' assay description and provide a guide for depositors on the submission of high quality data. ChEMBL's assays are typically structured with the overall aim, target, and method .  The ideal assay description is succinct but contains all the necessary information for easy interpretation by database u
Post a Comment
Read more