petermr's blog

I’ll now reply to the second part of Rich’s question

Rich Apodaca says:

April 4, 2009 at 4:49 pm (Edit)

Metallocenes? Axial Chirality? Apache/MIT/BSD License? OpenOffice? GitHub?

My current understanding is that C4W will be posted on CodePlex when we believe it’s in a reasonable state for community work. Good Open Source projects need clear release management and we haven’t yet addressed that, but we shall. I’m not very familiar with CodePlex but it tells me:

CodePlex is Microsoft’s open source project hosting web site. Start a new project, join an existing one, or download software created by the community.

and

Microsoft is hosting the CodePlex site solely as a web storage site as a service to the developer community

From WP I find:

CodePlex is an open source project hosting website from Microsoft. It allows shared development of open source software. Its features include wiki pages, source control based on Team Foundation Server but accessible using Subversion, discussion forums, issue tracking, project tagging, RSS support, statistics, and releases. Some of the available licenses are more restrictive than traditional open source licenses[1].

Choosing an Open Source licence is not trivial, and we haven’t yet chosen one. We’d welcome informed comment. The site offers, inter alia, Apache 2.0 and MsPL. JUMBO is Artistic because it allowed me to require forkers to rename and acknowledge. It’s compatible with MS Open philosophy. Much of the Blue Obelisk is LGPL which is probably compatible. Open Babel is GPL, probably incompatible with Microsoft licences.

Open Office? This probably relates to interoperability between the two systems. There are several aspects; some interoperates and some doesn’t. I’m visiting Peter Sefton in Toowoomba this month and we’ll probably find out what works and what doesn’t. Peter has managed to get CML molecules into ICE/ODT and interoperating with Open Office, but I don’t know the details. But as a rough guide I’d say:

• The C4W code is C# so only works in a environment that supports that. There are Open C# implementations, but I have no idea what they are like. I would doubt it suport the UI and graphics.
• WPF and XAML. Probably not interoperable. But I know of no high-quality graphics/UI system that is truly interoperable.
• customXML (where the CML is). I don’t know details, and I suspect they are somewhat hairy, but the CML itself is unaffected.
• OOXML and ODT. I suspect these are interoperable enough for most chemistry purposes.

GitHub. WP says:

GitHub is a web-based hosting service for projects that use the Git revision control system. It is written in Ruby on Rails by Logical Awesome developers Chris Wanstrath, PJ Hyett, and Tom Preston-Werner. GitHub offers both commercial plans and free accounts for open source projects. It is similar to Bitbucket which uses Mercurial.

I’m not sure how this is relevant. Currently we mainly use Subversion on Sourceforge, but Nico has mounted his ontology on BitBucket and I am sure he’ll tell us about it. We haven’t decided on the developer strategy yet but I would expect a single point of contact will work initially. I haven’t found MS’s TFS very cuddly yet and I prefer Subversion, but I am sure all these systems will develop.

Rich Apodaca is an original member of the Blue Obelisk and has developed his own chemical authoring tool (ChemWriter 299USD). He’s just posted the rather enigmatic comment…

Rich Apodaca says:

April 4, 2009 at 4:49 pm (Edit)

Metallocenes? Axial Chirality? Apache/MIT/BSD License? OpenOffice? GitHub?

I’m taking this to mean that he’s asking slightly tongue-in-cheek (a) about the power of C4W’s and CML’s representation of chemistry and (b) the openness and comunity aspects of C4W. It’s actually an excellent opportunity to follow those up. Here’s a recent post…

Language for Chemical Representation Part 2: Real-World Problems

Posted by Rich Apodaca 8 days ago

The last installment in this series discussed the limitations in today’s molecular languages and how FlexMol is designed to overcome them. Although these limitations are clearly present theoretically, what’s the practical effect likely to be?
For the last two years, a series of articles highlighting specific examples from the current chemical literature have appeared here. Variously titled “How would your cheminformatics tool do this?”, “Can your cheminformatics tool to this?”, and “Cheminformatics Puzzler”, each entry featured an article from a mainstream chemistry journal in which SMILES, Molfile, CML, and/or InChI would be incapable of faithfully representing a centerpiece structure. The examples are taken from well-read journals in synthetic organic, natural products, and medicinal chemistry.
The purpose was not to bash these languages, but rather point to an important common set of limitations among them – a kind of groupthink if you will.

and earlier

The fundamental problem with ‘standard’ molecular languages such as molfile, SMILES, InChI, and CML is their simplification of bonding and stereochemistry. Bonding is defined as an association between two atoms using two electrons. Stereochemistry is defined in terms of one or more chiral templates.
…
FlexMol takes a different approach. Bonding is defined in terms of systems of one or more pairs of atoms interacting with the cooperation of zero or more electrons. Stereochemistry is defined in terms planes passing through atom pair axes.
As we shall see, this flexible system enables the faithful, lossless representation of almost any chemical substance consisting of a single, well-defined molecular entity.

<p .
There's a fundamental misunderstanding here about the role of CMl (which is anything but group-think). CML addresses the semantics of chemistry. I could reply – in the same lighthearted vein:

Zeolites? Clathrates? Block co-Polymers? HEPES buffer? transition states? gaussian logfile output? cell dimensions? multiplets? eigenvectors?
and assert that CML can deal with all of them and ChemWriter cannot.

In fact CML can deal with any of the examples that Rich has mentioned in his article because it is (a) extensible (b) namespaced and (c) linked to ontologies. CML can add any properties to any of its primitives (atoms, bonds, etc.) It can define multicenter bonds and bonds between bonds. It has primitives for lines, planes, etc which should be sufficient for representing any of the geometry mentioned. JUMBO can do geometry and algebra on these if required. It has a primitive for electron. It also can hold 2D and 3D coordinates for atoms, so that it can represent the drawings of any of the species in Rich's diagrams.

That some of these CML primitives are not used in practice is because to be useful there needs to be agreement between two of more people. If Rich wishes people to use FlexMol then either they all have to use his software or other vendors have to install FlexMol readers and writers. If he can show me a groundswell of users of FlexMol and if it appears useful for them to convert to CML then I'd be happy to give some pointers.

What would emerge is a set of primitives and ontology terms that was FlexMol-specific - in CML we call call this a convention. There are already several conventions in CML - a typical example is a JSpecView convention for spectra. This requires that a spectrum contains data (it's perfectly reasonable to have an empty spectrum) so that JSpecView can display it. Another convention is CML-lite - a subset of primitives which are processable by default by C4W.

But because CML is semantic and because it uses ontologies it can hold a very wide range of chemistry. If a processor does not understand some of it, then it simply passes it through without loss. Whether this changes the semantics can be decided by the ontology and although that's at an early stage the basic infrastructure works.

I appreciate that for generations raised on FORTRAN-like formats (Mol) and implicit information (SMILES) that it will take time to migrate to XML-based ontology-driven chemistry. But it's the only way forward that can cover mainstream chemistry whether it be molecules, reactions, crystallography, nanotechnology, computation, spectra, physical properties and their measurement.

Because chemistry is a lot more than organic molecules drawn pictorially...

For many years I used to have a small patch of weeping scalp, but when I came to Cambridge I went to Ray’s barber shop in All Saints’ Passage – he was about 75 and got his politics from the Daily Mail but was otherwise an entertaining talker (you didn’t try to stop him). He told me that the patch was because I was an academic and I used my left brain more so I have this patch on my right scalp. I kept listening. So the way to get rid of it was Dead Sea Spa Magik Mineral Shampoo. It’s not cheap, but it’s not outrageous so I bought a bottle. And for whatever reason the patch disappeared and hasn’t come back.

I haven’t done a single blind trial on me and I couldn’t easily do a double blind one so it may be the placebo. However it works and I can just about afford it (I save by leaving longer times between haircuts…). Sadly Ray has upped shop…

Today I had to renew it and went to a well known supplier of food and neutraceuticals. Ever since reading Ben Goldacre’s Bad Science I now believe that almost all are worthless. (I used to believe antioxidants were useful but he showed from the literature that there were no proven benefits.)

So what’s in this Dead Sea stuff? I assumed it was a simple shampoo with masses of KCl and MgSO4 or something. Here goes:

• Harmonized water (TM) (Aqua Maris Sal) which I take to mean water from the Dead Sea. I do not know how it is harmonized or what that means. Maybe it’s homeopathic and you dilute with ultra concentrated salt solution.
• Sodium Coceth sulfate. Never heard of it but WP gives:
  “Sodium Coceth Sulfate is a semisynthetic detergent-like compound derived from fatty acids obtained from coconut oil, modified using ethylene oxide (oxirane). It is a milder foaming agent found in baby cleansers, gels, and cleaners.” Well, since I am supported to know all about polyethylenoxide this is useful knowledge.
• Cocamidopropyl betaine. Not sure what OSCAR/OPSIN would make of this name (I’ll have more on that later). I guess it’s coconut oil treated with a betaine. Yup, I was right (see WP). I don’t think OPSIN would get that yet but I expect Pubchem does. Look for yourself.
• Sodium beeswax (sic). I guessed this was saponification of beeswax and found:
  

  This derivative of Beeswax is derived through the saponficiation process, which involves the hydrolyzation of Beeswax in the presence of sodium in order to create soaps. It serves as a natural emulsifier and provides the same skin-caring properties as Cera Flava (Beeswax).

  I hope they have some hydroxide ions as well otherwise it could be quite fun.

• Zinc pyrithione. This is a well-characterized pure compound with a beautiful structure (5-coordinate Zn). WP says it’s antifungal and antidandruff so I just wonder whether that should be where I go next.
• Lauryl Pyrrlidone. It should be Lauryl pyrrolidone of course and for those of you – like me – who stop counting at capric acid it’s easier for OPSIN as 1-dodecylpyrrolidin-2-one.
• Sodium chloride. Wonder where they get that?
• Xanthum gum
• Fragrances, plants extracts, etc.

Ah, at the bottom it explains “harmonized”:

“Special de-ionised water with pure Dead Sea Minerals: Magnesium, Potassium, Calcium, Sodium as Bromides, Chlorides and Sulphates”

Isn’t the idea of specially deionising water and stuffing it full of concentrated salts enormous fun!

So, determining to get out as fast as possible and whizzing past the shelves that BadScience rightly attacks my eye was suddenly caught by
Methylsulfonylmethane
Since Daniel (to be explained later) has been improving OPSIN out of all recognition I can do this in my head and get MeSO2Me. It’s a solid (tablets). Never heard of it as a pharmaceutical. So when I get to the counter I ask what MSM does. The assistant looks it up and it cures arthritis, anxiety, and much else including with worms+parasites. (I can believe that – it looks as if it could be unpleasant although WP’s verdict is mostly harmless and probably worthless). Further information: “Active ingredient: sulfur”.

Oh dear. The Victorians used to give children brimstone (native sulfur). So I asked did they sell sulfur (they might). No, she said, but would I like some chondroitin sulfate..

Thanks I said, but no. I get my Sulfur in Calcium Sulfate from the water company – and they throw it in for free, every day.

I’m winding down on the library of the future theme – I am impatient to tell you about Chem4Word, whether and how we can rescue Cheminformatics from its current position as a pseudoscience, how to do language processing and textmining properly, and so on. But I owe my hosts at “The JISC” and Bodley thanks and the wider blogosphere some wind-down.

It’s been valuable to me and to some others I have spoken to. The event itself was great and brilliantly staged and Dicky, Helen and many others deserve great credit. (I will pass over the late night session with Dicky and Rachel, other than to say that after retiring at 0200 getting the 0651 back to London was a major effort of will.)

I loved the Twitter feed and SecondLife. The feed was instantaneous and had comments all round the world – search for #LOTF09 and PMR (http://search.twitter.com/search?q=%23lotf09+PMR) and you get lots of tweets – most of these while I was speaking and answering questions. It’s a new experience and I think it’s great (perhaps it was a good thing I couldn’t see the feeds during my talk – one was “Classic PMR”) . Here’s a few from the first pageL

tomroper: #LOTF09 Q: role of university presses in battle for scientific information? A PMR: publishing and learned organisations poor bedfellows

The SL was an experiment and it worked, within limits. The display was on except during talks, which I think was a pity and which should – in retrospect – be tried. I could see Jennifer/broniba/Akua_Inkpen but I couldn’t talk to her. Also any messages from SL were relayed through a human. But the streaming video in SL looked great from where I sat – the avatars were watching a live screen and we were watching live avatars. (I have an avatar but I haven’t worked out how to dress it – I need a lesson).

I need to clear up a misunderstanding which arose from my blogging – the library blogosphere felt I was attacking them. I was deliberately not, but I was giving them very . So I will briefly reply to a comment on John Dupuis’ blog

John Dupuis said…

I found it refreshing that although [PMR] started with the assumption that we’re all useless and dead, he did seem willing to at least listen and learn. Hopefully, we can listen and learn from him as well and see where we can improve.

At the end of the day, there’s a lot of mutual misunderstanding between scientists and librarians. It’s hard to know which is more of an issue: them not understanding us or us not understanding them.

PMR: If you read my earliest posts on this you will see that I did NOT assume librarians were “useless and dead” and I have never said that. I blogged #LOTF09 on the assumption I would get input from the library community. I know it was being read but even after nine days I got effectively no feedback. In bioscience, chemistry, information science, semantic web, open access, replies come within hours. I went into the FriendFeed and Twittersphere and picked up things like (paraphrased)  “we’ve written this report, why doesn’t PMR read it”, “why does PMR think that he can blog in our territory and expect a response”, “it’s PMR’s responsibility to approach us properly”, “he should apologize for criticizing us”.

So, getting no feedback, I turned up the outrage knob a bit and said

“I don’t blame the organizers (and I’m grateful to Dicky for sending the Ithaka report).  I’m left with the overwhelming impression that the community is now past caring about the future of the library. That’s essentially what Ithaka said 2-3 years ago – that ULibraries had to be visible and rebrand themselves.  They’re not and they aren’t.”

By community I meant the academic community as a whole, and I stand by what I said. I have tried to act as a messenger between  scientists and librarians. I have, I believe, been factually accurate. When JohnD says:

“It’s hard to know which is more of an issue: them not understanding us (ULibrarians)  or us not understanding them (scientists).”

I feel I am acting as a Cassandra or Jeremiah by saying that the issue is deadly clear. I am sorry to say the following, but it’s true.

Most scientists don’t care about (science) librarians. I talked last week with a very senior bioscientist – dean of science – editor of prestigious journal (which does its own data reposition, without help from libraries). I will not repeat what he said in detail becuase it will hurt too much, but a simple version is that libraries are a costly irrelevance and should be got rid of. I have no doubt that’s  a very widespread view in science.

There is no compulsion for the scientist to come to or understand the library. Scientists are already finding ways (Pubmed, EBI, Wikipedia, Nature Precedings, Wellcome Trust, etc.) to manage information without libraries. Scientific research manages hundreds of billions of dollars annually. ULibraries are topsliced, at least in part, from that.Topslicing is always an unpopular tax.

So I am trying to help libraries, rather than attacking them when I say that  it is up to them to approach scientists – and very rapidly – because that is where the money and the power is.

That’s the simple truth. It may be too late to do anything – I don’t know.

Robert Kiley has asked me to help garner comment on a potentially restrictive author-pays (or as may be funder-pays) licence. I’ll post him in full and add my own comment.

Open access licence: researcher opinion sought

• In the case of text-mining, User may incorporate individual words, concepts and quotes up to 100 words per matching sentence, whereas longer paragraphs of text and images cannot be used.
• Users may not create derivative works (as defined in the U.S. Copyright Act, 17 U.S.C. §101 et seq.) based upon the documents.

The Wellcome Trust is seeking input from the research community to help determine:

A) Whether such a licence would impact on your ability to re-use and re-purpose this open access content.

B) If so, please give some examples of research activities that would be limited by this licence.

If you would like to respond to this issue, please use the comment function below or send an email to r dot kiley at wellcome dot ac dot uk.

PMR: The first thing to remember is that funder-pays Open Access is – in large part – a business model. The funder wishes a good – universal access to the information – and is prepared to pay the provider an appropriate sum. There is, possibly, a balance between the fee and the extent of the service provided (although personally I am an absolutist). IOW it could be that the publisher says “we shan’t charge you very much and this makes author-pays access affordable to those who normally couldn’t. But understand that we cannot give so much in return.” I doubt this is the case here – I do not know and shall not guess the society but I suspect that the fees are comparable with normal author-pays.

A major problem is “what is a fair price for author-pays”. I have complete trust that Wellcome has thought very hard about this, because it sets the scene for the market. If the society is properly run for the benfit of its members then they should be able to determine whether the fee covers the cost of processing (after all these are non-profit orgs). If the books are closed and the fee is large then there is a suspicion that revenue is used for other purposes or that there is no drive towards effficiency.

It’s complicated because the society probably gets most of its revenue from closed access publications – they don’t stop coming and the subscriptions don’t stop either. So the society should lower its subscriptions pro rata to the author-pays income. Is there any evidence for this?

The wording of the licence implies – though I would hate to impugn any society publisher – that the society (or its senior officers) cares more abouyt income than about the furtherance of the scientific domain in which the community has invested its trust. A scientific society, especially in the chemistry/bioscience area should actively welcome text-mining as this is a major tool for extracting science from textual publications. I was sorry, for example, when Nature (for whom I have many regards) created the OTMI (open text mining initiative) where they exposed the text of the paper but jumbled the the order of the sentences. This obfuscation of science sits badly on a learned organization (even a for-profit one).

Our group text-and data-mines chemistry and bioscience. In some cases this is a worthy activity because running text is the way that bioscience is expressed. In others, such as chemical data, it’s because the authors, publishers, readers, don’t yet understand the value of semantics and ontologies (Chem4Word addresses that problem at the authoring stage).

What worries me about the licence is that the society wishes to restrict the use of semantic information. Why? Possibly because it has an interest in redistribtuing its own semantic version of the work or an aggregate at a higher cost. If so we are seeing yet again deliberate restriction in the progress of science.

Finally I really discourage the use of NC licences. They are difficult to manage, have complexities in mashups and derivative works. It’s a pity that the proponents of Gold-OA haven’t universally insisted on CC-BY – there are many who do like SPARC , BMC, PLoS, etc. Anything less is a less valuable good and should not carry the same fee.

I understand that restrictions and politics in the US have made it difficult to get full CC-BY for the NIH policy (“eyeballs are sufficient”) but we must not let this creep.

I took several months off blogging – completely. Why? Because I was concentrating on Chem4Word – a semantic chemical authoring tool sponsored by and jointly built with Microsoft External Research in Redmond. In this and subsequent posts I’ll now tell you about it and what hopes we have for it. Yesterday – and I’ll explain why in later posts – was a breakthrough day for us and I felt I could start to tell the world about it.

When we started I had no idea of the amount of mental and emotional effort I was going to have to put into this. We started about 2 years ago when Tony Hey and Ley Dirks invited some academic chemical informatics folk to Redmond to plan what has started as eChemistry and has now turned out to be OREChem. I’ll be blogging about this regularly elsewhere. But Tony and Lee also asked me if I would like our group and his to jointly develop a chemical drawing tool (I think that’s the phrase used then) for Word.

Why our group? Why not a commercial software company with lots of experience, with customers, with proven algorithms, etc. I can’t put words into his mouth but remember that Tony ran the UK eScience program and got a view of many disciplines – medicine, earth, environment, transport, oil, and particularly bioscience. These disciplines were applying the new tools of the distributed web, looking to share semantic information and prepare for the multi-party clouds that we are now seeing. He visited the ACS-CINF session to see what the state of the art was, and I’ll have to leave him to say what he concluded.

Microsoft – through the strenuous efforts of Jean Paoli – was a very early adopter of XML and so it was natural that Tony should look to CML (Chemical Markup Language) as the basic of our project. We’ve started from day zero with an XML data model, which has an amazing number of benefits. It supports side-effect-free programming (ably steered by Jim Downing) and can take advantage of the incredibly powerful LINQ system in dotNet (it’s really “.NET”) but it’s very easy to miss the dot.

So we started with a very fluid high-level plan and at the same time started the contractual negotiations. For those of you who don’t know, *all* large companies take a long time to finalise contracts on the first occasion. (And so do many small ones…). And Universities aren’t always quick. So the contract is measured in hundreds of days negotation – many of these anticipating the problems we might run into. (In retrospect – no surprise – a lot of this seems very periperal or even obsolete.

The basis of the chemistry was a CML engine. I’d already written JUMBO (Java Universal Molecular browser for Objects), which is neither Universal and no longer a browser – but this is converted into dotNUMBO (in C#).I’d hoped that we could autoconvert the software and I’m very glad we didn’t try. Although all lines of code have been typed in blood the design is much cleaner and it’s much smaller. And we are startiung with a subset of chemistry.

The effort has been incredible. I think I and Jim are nominally 10% of our time, but that’s >10% of 162 hours/week at times. Over the last 2-3 months we’ve taken to having daily telcons – sometimes short and sometimes cancelled – but often over an hour. And this effort has been made by the MS people as well – we’ve had many visits from Lee, Alex, Savas, and also a week’s stay from JimM. I think it’s taken over some of their lives as well….

It’s been a bigger project than any of us thought (I think). It’s been run on project management system (TFS) where we all invent scenarios and all get (multiple) tickets for tasks. Many of these are past their sell-by date. At times they hung over me (at least) as a cloud of doom, showing how slowly the project was going. Parts of milestones were missed. Scenarios moved to later milestones, etc.

But we set ourselves two deadlines which could not be moved. This is often a good way of helping to ensure communal vision and prioritisation. The deadlines were

• yesterday (the Microsoft External Research symposium, where we would present to a wider range of academics and MS staff
• end of April – BioIT is Boston – a large event with many commercial vendors and purchasers.

We made it yesterday with 10 minutes to spare. It was great. More – much more – later…