Skip to main content

Drug Side Effect Prediction and Validation


There's a paper just published in Nature getting a lot of coverage on the internet at the moment from Novartis/UCSF, and for good reason - but as the cartoon above states, it will probably have less impact than news on Justin Bieber's new haircut, or the latest handbags from Christian Lacroix. It uses the SEA target prediction method, trained using ChEMBL bioactivity data in order to predict new targets (and then by association side effects) for existing drugs. These are then experimentally tested, and the results confirmed in a number of cases - this experimental validation is clearly complex and expensive, so it is great news that in silico methods can start to generate realistic and testable hypotheses for adverse drug reactions (there are also positive side effects too, and these are pretty interesting to look for using these methods as well).

The use of SEA as the target prediction method was inevitable given the authors involved, but following up on some presentations at this springs National ACS meeting in San Diego. There would also seem to be clear benefits in including other methods into linking a compound to a target - nearest neighbour using simple Tanimoto measures, and naive Bayes/ECFPP type approaches. The advantage of the SEA approach is that it seems to generalise better (sorry I can't remember who gave the talk on this), and so probably can make more comprehensive/complete predictions, and be less tied to the training data (in this case ChEMBL) - however as databases grow, these predictions will get a lot better. There will also be big improvements possible if other data adopts the same basic data model as ChEMBL (or something like the services in OpenPHACTS), so methods can pool across different data sources, including proprietary in-house data.

There are probably papers being written right now about a tournament/consensus multi-method approach to target prediction using an ensemble of the above methods. (If such a paper uses random forests, and I get asked to review it, it will be carefully stored in /dev/null) ;)

So some things I think are useful improvements to this sort of approach.

1) Inclusion of the functional assays from ChEMBL in predictions (i.e. don't tie oneself to a simple molecular target assay). The big problem here though is that although pooling of target bioassay data is straightforward - pooling/clustering of functional data is not.
2) Where do you set affinity thresholds, and how do the affinities related to the pharmacodyamics of the side-effects. My view is that there will be some interesting analyses of ChEMBL that maybe, just maybe, allow one to address this issue. Remember, we know quite a lot about the exposure of the human body, to  a given drug at a given dose level...
3) Consideration of (active) metabolites. It's pretty straightforward now to predict structures of likely metabolites (not at a quantitative level though) and this may be useful in drugs that are extensively metabolised in vivo.

Anyway, finish off with some eye-candy, a picture from the paper (hopefully allowed under fair use!).


And here's a reference to the paper, in good old Bell AT&T labs refer format - Mendeley-Schmendeley as my mother used to say when I was a boy.

%T Large-scale prediction and testing of drug activity on side-effect targets
%A E. Lounkine
%A M.J. Keiser
%A S. Whitebread
%A D. Mikhailov
%A J. Hamon
%A J.L. Jenkins
%A P. Lavan
%A E. Weber
%A A.K. Doak
%A S. Côté
%A B.K. Shoichet
%A L. Urban
%J Nature
%D 2012
%O doi:10.1038/nature11159

Comments

Popular posts from this blog

Release of ChEMBL 33

We are pleased to announce the release of ChEMBL 33! This fresh release comes with a few new data soures and also some new features: we added bioactivity data for understudied SLC targets from the RESOLUTE project and included a flag for Natural Products and for Chemical Probes. An annotation for the ACTION_TYPE of a measurement was included for approx. 270 K bioactivities. We also time-stamped every document in ChEMBL with their CREATION_DATE! Have fun playing around with ChEMBL 33 over the summer and please send feedback via chembl-help@ebi.ac.uk .   ChEMBL database version ChEMBL 33 release notes ___________________________________________ # This version of the database, prepared on 31/05/2023 contains:      2,399,743 compounds (of which 2,372,674 have mol files)      3,051,613 compound records (non-unique compounds)        20,334,684 activities         1,610,596 assays      15,398 targets      88,630 documents BioAssay Data Sources:    Number Assays:    Number

This Python InChI Key resolver will blow your mind

This scientific clickbait title introduces our promised blog post about the integration of UniChem into our ChEMBL python client. UniChem is a very important resource, as it contains information about 134 million (and counting) unique compound structures and cross references between various chemistry resources. Since UniChem is developed in-house and provides its own web services , we thought it would make sense to integrate it with our python client library . Before we present a systematic translation between raw HTTP calls described in the UniChem API documentation and client calls, let us provide some preliminary information: In order to install the client, you should use pip : pip install -U chembl_webresource_client Once you have it installed, you can import the unichem module: from chembl_webresource_client.unichem import unichem_client as unichem OK, so how to resolve an InChI Key to InChI string? It's very simple: Of course in order to reso

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

Chemistry and Nature

  As the Great Big Green Week (UK) draws to a close, so does EMBL-EBI’s own Sustainability week. The Wellcome Genome Campus held events in the areas of recycling, energy use, and biodiversity. The ChEMBL team was keen to get involved and we developed our own Nature Trail event highlighting some of the bioactive compounds from the flora and fauna found on-site, and elsewhere. Our favourite examples include the sensation of mint and chilli and the glorious smell of rain! The full Nature Trail can be made available for external Public Engagement events upon request . Databases, such as ChEMBL , are large stores of structured data, including genetic, biological, and chemistry data for life sciences research. Data on the natural world is often held by wildlife organisations; this can be used to research biodiversity and species decline. Various Citizen Science initiatives mean that everyone can get involved in submitting nature records. So why not join in with the Butterfly Conservation’s B

Drug warning update: withdrawn drugs and drugs that carry a black box warning

The drug warning information in ChEMBL has been updated for version 32. In particular, the withdrawn drug data has been fully reviewed and, to assist the manual curation process, our rules have been updated, clarified and formally written.  In ChEMBL, a withdrawn drug is an approved drug (ie Phase 4) that has subsequently been withdrawn for toxicity reasons. For example, a drug is assigned as 'withdrawn' if: All doses are withdrawn (and not just the highest dose). The drug is withdrawn for all populations (and not just infants). The drugs is withdrawn for all indications.  Any drug withdrawn for a lack of evidence of efficacy is not included.  Any drug withdrawn for drug-drug interactions is included if it is a safety-related withdrawal. A regulatory body (e.g. EMA, FDA) is the preferred source of information for the withdrawn status. The withdrawn status is mapped to an individual drug form (e.g. a parent (salt-stripped) or salt drug form within a family of compound structures