The ChEMBL-og

Have you ever wondered which compound is the most popular in ChEMBL? And by popular I don't mean the one which cracks the best jokes at dinner parties; I mean the compound with the largest number of structural analogues or nearest neighbours (NNs). This number also gives an indication of the sparsity or density of the chemical space around a compound and is a useful concept during hit expansion and lead optimisation.  This number of course depends on the fingerprint, the hashing and folding parameters, the similarity coefficient and the threshold. So let's say 2048-bit RDKit Morgan fingerprints with a radius of 2 or 3 (equivalent to ECFP_4 or ECFP_6) and Tanimoto threshold of 0.5. Why so low threshold? For an explanation, see here and here . To calculate this compound 'popularity', one would need to calculate the full similarity matrix of the 1.4M compounds in ChEMBL. This used to be prohibitively computationally expensive just a few years ago; nowadays,...

Following the success of our ChEMBL Bot , there is now a new faithful bot out there which answers to the name  @MalariaSARLit and is looking for new followers. Its job is to tenaciously monitor PubMed for new malaria-related publications, score them according to our ChEMBL-likeness score and tweet a ChEMBL-like one daily at noon GMT. Followers of the bot will get a  free and reliable antimalarial SAR paper alert every day in their twitter feed.   George (NKOTB fan)

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 e...

For those who forgot to register, this is a gentle reminder for the 2nd RDKit User Group Meeting.  The  meeting  will take place  October 2nd-4th  here the  Genome Campus in Hinxton, UK.  We're using a different format for the meeting this year: Days 1 and 2:  Talks, lightning talks, roundtable(s), discussion, and something new:  talktorials!  Talktorials are somewhere between a talk and a tutorial, they cover something interesting done with the RDKit and include the code used to do the work. During the presentation you'll give an overview of what you did and also show the pieces of the code that are central to the work. The idea is to mix the science up with the tutorial aspects. Day 3  will be the first ever RDKit sprint: those who choose to stay will spend an intense day working in small groups to produce useful artifacts: new bits of code, KNIME nodes, KNIME workflows, tutorials, documentation, IPython notebooks,  e...

There are only two issues with website implementation - 'interface' and 'content'. We have a project here at the EMBL-EBI connected with interface and UI principles, and they have a blog, here is a link to that blog .

ChEMBL small molecule-protein interactions are now available in PSI-MI TAB and XML formats, thanks to the Proteomics Services group. This dataset includes ChEMBL interactions identified via binding assays with IC50/Ki/EC50/Kd values below 10uM - just under 500,000 interactions in total (with negative/weaker interactions also included in the XML export). The data can be accessed via the PSICQUIC project (Proteomics Standards Initiative Common QUery InterfaCe), which provides programmatic access to a wide range of molecular interaction databases via SOAP and REST web services. For example, this URL retrieves all ChEMBL interactions relating to imatinib (Gleevec). Mail us if you need more info.

I am on holiday today - sort of. Went to Borders for a Starbucks (product placement hopefully pays well), and while queuing for my Orange Mocha Frappuccino! , I caught sight of the O'Reilly books ; one stood out from the crowd - Programming Collective Intelligence . It looks a very cool collection of code ( Python ) implementing a whole variety of data analysis/machine learning techniques and routines to build smarter, more responsive and adaptive web 2.0 applications. Skimming the pages while I had my caffeine speedball led me to spend my cash. %A Toby Segaran %T Programming Collective Intelligence %I O'Reilly %D 2007 %O ISBN 978-0596529321

Just downloaded and synced up Papers for the iPhone from mekentosj . What a great little app, beautiful interface, very snappy performance, and allows the carrying of a whole bunch of literature in your pocket, and also searching/downloads from your handheld. What more could a hipster mobile scientist want? (Apart from good 3G coverage, 64GB of memory, free Wi-Fi everywhere, and free journal access, of course).

This was the first proper programming book that I studied, and it is an old one too, written before the web, web-services, and networking - good old fashioned UNIX programming computation. Few words, lots of concepts, and merits revisiting now and then - the prose in Kernighan's books is just excellent, perfect pace, and combines advice with examples of little code snippets. Probably the best thing about the book (for me) is the use of ratfor - a derivative of Fortran that looks a lot like C, and has some of the best elements of both languages. It is also incredibly quick to code in. I have just downloaded some updated versions of ratfor for my mac, so expect some pretty unusual looking Open Source protein analysis tools anyday soon! Anyway, although ratfor is used, for almost everyone, the code there will just be pseudocode for implementation and inspiration of code in a more fashionable language. If anything, using a non-current language forces thinking more deeply about the...

You know what it's like - you have a deadline, stuff that's really important to do, you get your work area ready, and then you browse your bookcase for something interesting. Four hours later, it's time to do something else. Well that was yesterday, and I spent those hours re-reading this old classic. Fairly recently released in a reprinted, and cheaper form. In my opinion, this is one of the best books in sequence comparison, it is full of interesting ideas, has good coverage of related fields of computer science, and a coverage of the algorithms that are deep enough to allow you to go away and start messing around with code. %D 2000 %E David Sankoff & Joseph Kruskal %T Time Warps, String Edits, and Macromolecules: The Theory and Practice of Sequence Comparison %I Cambridge University Press %O ISBN 978-1575862170

All this travelling is making me read. This was a book I picked up over the summer, but have only recently properly read. It is a classic in the open source field, and is a very lucid discussion of the culture of open source projects. It is excellent in quite a few places, and my copy is now dog-eared where I thought at the time that I must remember some point or other. Many things I simply had not though of before, in particular the multifaceted discussions of 'free' vs commercial were particularly interesting. One definite down side was that I felt like I was 'l33t' after I had finished it, and tried to bore my kids with my new insights - I guess this is may be the modern equivalent of you dad dancing at a disco. %T The Cathedral And The Bazaar %A Eric Raymond %I O'Reilly Media, Inc. %D 2001 %O ISBN 978-0596001087

This year, Santa Claus delivered a book (I always thought he lived at the North Pole, but clearly he is now based in The Amazon) that I had seen advertised a few places - 'Information Dashboard Design: The Effective Visual Communication Of Data', by Stephen Few. It is one of those compelling books that although the fundamental message is simple and arguably obvious, it is nonetheless a delight to read, and I learnt a lot from it. The basic theme of the book is in the necessary features and design of intuitive interfaces, and in particular those that need to display quantitative and comparative numerical data. My first contact with books of this type were with the classics of Edward Tufte, which remain timeless, but are complemented by this book addressing HCI issues. %D 2008 %A Stephen Few %T Information Dashboard Design: The Effective Visual Communication Of Data %I O'Reilly %O ISBN 978-0596100162