Competition Time - Win a Raspberry Pi with ChEMBL - chempi

Here's a free to enter competition for a brand new, fully working raspberry pi running the brand new chempi implementation. It includes everything you need to get started at home with ChEMBL - a sort of in silico Breaking Bad maybe (hopefully not, thinking about it). It includes everything you need, with the exception of a power supply and ethernet cable.



We have run out of our creative juices, and cannot think of a suitable poem to mark the release of chempi - so the competition is for you to finish a limerick for us, starting with the line.



There once was a hacker with chempi....



Entries must be posted in the comments section. Obscene or defamatory entries will be removed (all comments are moderated, so it may take a few hours for you entry to appear, so do not repost twenty times!). We haven't really decided how to pronounce chempi (with a hard 'k' start or a soft 'sh' start, just as with ChEMBL, both are used in the wild; and also does it rhyme with scampi, or the irrational number pi?). All entries will be assumed to be made under CC-BY licensing. The competition will be open until noon GMT on Sunday 10th November 2013.



Entires will be judged for compliance to a standard limerick format, outrageous rhymes with chempi, gratuitous chemistry references, and finally humour.



The judges decision (i.e. mine) is final. The winning entry will be published on the ChEMBL-og.



jpo



PS Before I get asked, the competition is not open to members of the ChEMBL group, or extended family members of the ChEMBL group.

Document Similarity in ChEMBL - 1

Many of you will have noticed a new section on the ChEMBL interface, specifically at the Document Report Card page, called Related Documents. It consists of a table listing the links for up to 5 other ChEMBL documents (i.e. publications aka papers) that are scored to be the most similar to the one featured in the report card. Here's an example. 

How does this work? There are examples of related documents sections online, e.g. in PubMed or in various journal publishers' websites. Document 'related-ness' or similarity can be assessed by comparing MeSH keywords or by clustering documents using TF-IDF weighted term vectors. Fortunately, ChEMBL puts a lot of effort in manually extracting and curating the compounds and biological targets from publications, so why not using these as descriptors to assess document similarity instead - as far as we know this is the first time this approach has been implemented?

So, here's how it works:

Firstly, for each document in ChEMBL, its list of references is retrieved using the excellent EuropePMC web services. By considering documents as nodes which are connected with an edge if one paper cites the other, a directed graph structure emerges. By doing this for all ~50K documents in ChEMBL, you get the massive graph illustrated above in Cytoscape. As a bonus, by measuring the in- and out- degree of the nodes, one could check which are the most cited papers in ChEMBL - but that's the topic of another blog post. This graph could be further annotated with protein target families, authors and institutions, as it has been elegantly done here.

Moving on, once a relationship between two documents is established, we need a way to quantify their similarity. As hinted above, we used the normalised overlap of compounds and targets reported in the two documents. This is done using the classic Tanimoto coefficient, so if doc A reports compounds (1,2,3) and doc B reports compounds (3,4,5), their compound Tanimoto similarity T is 1/5 or 0.2. Exactly the same applies for the target-based document similarity. The composite score we use to rank docs in the Related Documents section is simply the maximum of the two individual ones.

What does all that mean in practice? It means that 2 papers are listed as similar if they their reported compounds or biological targets overlap significantly (and one cites the other). For example, papers with follow-up experiments on the same candidate drug will be deemed similar, e.g. this one. The same will apply to two papers that involve kinase panel screening assays. A desirable side-effect is that by following the links, the tenacious user may traverse the whole graph displayed above! 

George & Mark 

ChEMBL_17 Released

We are pleased to announce the release of ChEMBL_17. This version of the database, prepared on 29th August 2013 contains:

• 1,519,640 compound records


• 1,324,941 compounds (of which 1,318,187 have mol files)


• 12,077,491 activities


• 734,201 assays


• 9,356 targets


• 51,277 documents

You can download the data from the ChEMBL FTP site. For more information please read the release notes.



Data changes since the last release:



Drug mechanism of action



For all FDA-approved drugs, information regarding the mechanism of action and associated efficacy targets has been curated from primary sources, such as literature and drug prescribing information. Targets have only been included for a drug if a) the drug is believed to interact directly with the target and b) there is evidence that this interaction contributes towards the efficacy of that drug in the indication(s) for which it is approved.



Metal-containing compounds



Structures for around 3200 metal-containing compounds have been removed from the database (though the bioactivity and other information for these compounds is retained). For more information, please see the previous blog posts: http://chembl.blogspot.co.uk/2013/08/removal-of-metal-containing-compounds.html



New data sets



Several new deposited/extracted data sets have also been included in the latest release: two deposited data sets from GlaxoSmithKine for Ghrelin receptor agonists and Motilin receptor agonists, a data set of the results of screening the MMV Malaria Box compound collection for activity against Schistosoma mansoni, two data sets screening the GSK PKIS compound collection for inhibition of luciferase activity, and finally pathology data from the Open TG-GATES project.



Interface changes since the last release:



Browse Drug Targets tab



A new tab has been created to show the new mechanism of action information for FDA approved drugs together with the references from which the information was obtained, and links to the relevant drug/target report card pages.

Document Report Card



A new table has been added to the document report card, showing other ChEMBL documents that are related to the current document. Pair-wise document similarity is assessed by two components. The first component is defined by whether a document cites or is referenced by the other. The second component is defined by the amount of overlap between the compounds and biological targets reported in the two respective documents. This overlap is quantified by the Tanimoto coefficient. Documents with the highest Tanimoto similarity scores to the query document are listed in this section. For example, the following page shows 5 additional ChEMBL documents that are deemed similar to the paper currently being viewed.

Database changes since the last release:



A number of new tables have been added to store the drug mechanism of action information (please see release notes and schema documentation for full details). In addition, a number of minor changes have been made to existing tables:



The PROTEIN_FAMILY_CLASSIFICATION table has been deprecated and replaced by a new hierarchical version: PROTEIN_CLASSIFICATION.



The MOLREGNO field has been removed from the ATC_CLASSIFICATION table and moved to a new mapping table: MOLECULE_ATC_CLASSIFICATION.



The MOLFORMULA field has been moved from the COMPOUND_STRUCTURES table to the COMPOUND_PROPERTIES table (and renamed).





The ChEMBL Team