Skip to main content

Ligand-based target predictions in ChEMBL



In case you haven't noticed, ChEMBL_18 has arrived. As usual, it brings new additions, improvements and enhancements both on the data/annotation, as well as on the interface. One of the new features is the target predictions for small molecule drugs. If you go to the compound report card for such a drug, say imatinib or cabozantinib, and scroll down towards the bottom of the page, you'll see two tables with predicted single-protein targets, corresponding to the two models that we used for the predictions. 


 - So what are these models and how were they generated? 

They belong to the family of the so-called ligand-based target prediction methods. That means that the models are trained using ligand information only. Specifically, the model learns what substructural features (encoded as fingerprints) of ligands correlate with activity against a certain target and assign a score to each of these features. Given a new molecule with a new set of features, the model sums the individual feature scores for all the targets and comes up with a sorted list of likely targets with the highest scores. Ligand-based target prediction methods have been quite popular over the last years as they have been proved useful for target-deconvolution and mode-of-action prediction of phenotypic hits / orphan actives. See here for an example of such an approach and here for a comprehensive review.


 - OK, and how where they generated?

As usual, it all started with a carefully selected subset of ChEMBL_18 data containing pairs of compounds and single-protein targets. We used two activity cut-offs, namely 1uM and a more relaxed 10uM, which correspond to two models trained on bioactivity data against 1028 and 1244 targets respectively. KNIME and pandas were used for the data pre-processing. Morgan fingerprints (radius=2) were calculated using RDKit and then used to train a multinomial Naive Bayesian multi-category scikit-learn model. These models then were used to predict targets for the small molecule drugs as mentioned above. 


 - Any validation? 

Besides more trivial property predictions such as logP/logD, this is the first time ChEMBL hosts non experimental/measured data - so this is a big deal and we wanted to try and do this right. First of all, we did a 5-fold stratified cross-validation. But how do you assess a model with a many-to-many relationship between items (compounds) and categories (targets)? For each compound in each of the 5 20% test sets, we got the top 10 ranked predictions. We then checked whether these predictions agree with the known targets for that compound. Ideally, the known target should be correctly predicted at the 1st position of the ranked list, otherwise at the 2nd position, the 3rd and so on. By aggregating over all compounds of all test sets, you get this pie chart:


This means that a known target is correctly predicted by the model at the first attempt (Position 1 in the list of predicted targets) in ~69% of the cases. Actually, only 9% of compounds in the test sets had completely mis-predicted known targets within the top 10 predictions list (Found above 10). 

This is related to precision but what about recall of know targets? here's another chart:



This means that, on average, by considering the top 10 most likely target predictions (<1% of the target pool), the model can correctly predict around ~89% of a compound's known single protein targets. 

Finally, we compared the new open source approach (right) to an established one generated with a commercial workflow environment software (left) using the same data and very similar descriptors:


If you manage to ignore for a moment the slightly different colour coding, you'll see that their predictive performance is pretty much equivalent.

 - It all sounds good, but can I get predictions for my own compounds?

We could provide the models and examples in IPython Notebook on how to use these on another blog post that will follow soon. There are also plans for a publicly available target prediction web service, something like SMILES to predicted targets. Actually, if you would be interested in this, or if you have any feedback or suggestions for the target prediction functionality, let us know

George

Comments

Unknown said…
Very nice post, cheers!
Unknown said…
Any thoughts on the domain of validity in chemical space of these models? Do you expect them to work well across all of chembl, and if not can you specify what compounds they will fail on?
Unknown said…
Thank You for the very interesting work! I have some questions. First of all, i don't quite understand your validation technique. For example: a compound has 3 targets. Target 1 was found at the first position; target 2 was found at the second position and target 3 was not found in top 10 list of predictions. What did you do exactly in similar cases? Second, how many compounds are there in your training set?

Popular posts from this blog

A python client for accessing ChEMBL web services

Motivation The CheMBL Web Services provide simple reliable programmatic access to the data stored in ChEMBL database. RESTful API approaches are quite easy to master in most languages but still require writing a few lines of code. Additionally, it can be a challenging task to write a nontrivial application using REST without any examples. These factors were the motivation for us to write a small client library for accessing web services from Python. Why Python? We choose this language because Python has become extremely popular (and still growing in use) in scientific applications; there are several Open Source chemical toolkits available in this language, and so the wealth of ChEMBL resources and functionality of those toolkits can be easily combined. Moreover, Python is a very web-friendly language and we wanted to show how easy complex resource acquisition can be expressed in Python. Reinventing the wheel? There are already some libraries providing access to ChEMBL d

ChEMBL 29 Released

  We are pleased to announce the release of ChEMBL 29. This version of the database, prepared on 01/07/2021 contains: 2,703,543 compound records 2,105,464 compounds (of which 2,084,724 have mol files) 18,635,916 activities 1,383,553 assays 14,554 targets 81,544 documents Data can be downloaded from the ChEMBL FTP site:   https://ftp.ebi.ac.uk/pub/databases/chembl/ChEMBLdb/releases/chembl_29 .  Please see ChEMBL_29 release notes for full details of all changes in this release: https://ftp.ebi.ac.uk/pub/databases/chembl/ChEMBLdb/releases/chembl_29/chembl_29_release_notes.txt New Deposited Datasets EUbOPEN Chemogenomic Library (src_id = 55, ChEMBL Document IDs CHEMBL4649982-CHEMBL4649998): The EUbOPEN consortium is an Innovative Medicines Initiative (IMI) funded project to enable and unlock biology in the open. The aims of the project are to assemble an open access chemogenomic library comprising about 5,000 well annotated compounds covering roughly 1,000 different proteins, to synthesiz

Using ChEMBL web services via proxy.

It is common practice for organizations and companies to make use of proxy servers to connect to services outside their network. This can cause problems for users of the ChEMBL web services who sit behind a proxy server. So to help those users who have asked, we provide the following quick guide, which demonstrates how to access ChEMBL web services via a proxy. Most software libraries respect proxy settings from environmental variables. You can set the proxy variable once, normally HTTP_PROXY and then use that variable to set other related proxy environment variables: Or if you have different proxies responsible for different protocols: On Windows, this would be: If you are accessing the ChEMBL web services programmatically and you prefer not to clutter your environment, you can consider adding the proxy settings to your scripts. Here are some python based recipes: 1. Official ChEMBL client library If you are working in a python based environment, we recommend

LSH-based similarity search in MongoDB is faster than postgres cartridge.

TL;DR: In his excellent blog post , Matt Swain described the implementation of compound similarity searches in MongoDB . Unfortunately, Matt's approach had suboptimal ( polynomial ) time complexity with respect to decreasing similarity thresholds, which renders unsuitable for production environments. In this article, we improve on the method by enhancing it with Locality Sensitive Hashing algorithm, which significantly reduces query time and outperforms RDKit PostgreSQL cartridge . myChEMBL 21 - NoSQL edition    Given that NoSQL technologies applied to computational chemistry and cheminformatics are gaining traction and popularity, we decided to include a taster in future myChEMBL releases. Two especially appealing technologies are Neo4j and MongoDB . The former is a graph database and the latter is a BSON document storage. We would like to provide IPython notebook -based tutorials explaining how to use this software to deal with common cheminformatics p

Using autoencoders for molecule generation

Some time ago we found the following paper https://arxiv.org/abs/1610.02415 so we decided to take a look at it and train the described model using ChEMBL. Lucky us, we also found two open source implementations of the model; the original authors one https://github.com/HIPS/molecule-autoencoder and https://github.com/maxhodak/keras-molecules . We decided to rely on the last one as the original author states that it might be easier to have greater success using it. What is the paper about? It describes how molecules can be generated and specifically designed using autoencoders. First of all we are going to give some simple and not very technical introduction for those that are not familiar with autoencoders and then go through a ipython notebook showing few examples of how to use it. Autoencoder introduction Autoencoders are one of the many different and popular unsupervised deep learning algorithms used nowadays for many different fields and purposes. These work wi