ChemSpider News

ChemSpider News Switches Off. ChemConnector Switches On

ChemSpiderMan — Wed, 02 Jan 2008 19:15:43 +0000

For the past 8 months I’ve been running two blogs associated with ChemSpider…the ChemSpider blog and this blog, ChemSpider News. These had different intents – the ChemSpider blog was to cover the general directions, vision, activities, challenges and community conversations about the ChemSpider service. ChemSpider News was to cover details of new functionality as it was rolled out. What has happened is that the postings on this blog regarding new functionality have been missed by the readers of the other blog as a result of the readership being different (and different subscriptions to Feedburner being in place). This has not been good for the ChemSpider users since they have not been getting the latest and greatest news about new functionality. It’s a great shame to deliver new functionality and not have people use it for weeks on end because they don’t know about it.What I’ve also been posting on the ChemSpider blog are my own personal views regarding a whole variety of issues…none of them ChemSpider related. For example, my views on vaccines, on fluoride, on drugs, and on my views of various research undertakings. Recently John Doe took me to task in a recent response on the ChemSpider blog. While I vehemently disagree with his comment regarding the reasons behind my posting I have decided to separate news regarding ChemSpider and its mission, direction and new functionality from my own personal commentaries, pleas, musings and, once in a while, thigh-slapping fun exchanges (especially the ones about erectile dysfunction drugs and how the drug industry can learn from the golfing industry).

So, this site, while it will remain online for past references , will be discontinued in terms of new postings and all new functionality comments will be posted to the ChemSpider Blog. The ChemSpider blog will become just that…all about ChemSpider.

Regarding ChemConnector and it’s associated Blog. That blog will be a communication vehicle for the website www.chemconnector.com, to go online shortly. As many of you will know me from the ChemSpider blog you will likely be aware that I took a sabbatical after 10 years working a company and have taken a few months down. For the time-being my intention is to spend some time consulting for groups and organizations interested in utilizing my skills and growing the business of ChemConnector – Connecting Scientists to Problem Solvers. This, of course, in parallel to the ongoing work on ChemSpider. More on this will come shortly and will be exposed on the site.

Searching InChIKeys by Connectivities Only – With and Without Stereo

ChemSpiderMan — Wed, 12 Dec 2007 04:43:25 +0000

At the prompting of one of our users, and a clever idea it was too, we have implemented the ability to use the “split-nature” of the InChIKey to facilitate searches by connectivity only, and thereby ignoring the stereo layer for example. Wikipedia declares: InChIKeys consist of 14 characters resulting from a hash of the connectivity information of the InChI, followed by a hyphen, followed by 8 characters resulting from a hash of the remaining layers of the InChI, followed by a single character indication the version of InChI used, followed by a single checksum character.

So, BQJCRHHNABKAKU-XKUOQXLYBY is for morphine .

Now, if you click on the first part of the string you will get a connectivity structure search of Google.

and………… if you click on the second part of the string you will get an exact structure search of Google.

We will add the ability to search within ChemSpider or via Google or other search engines shortly.

Enhanced Searching for Mass Spectrometry

ChemSpiderMan — Wed, 12 Dec 2007 04:14:05 +0000

When we first set up the screens for searching the properties associated with a structure we set in place the ability to input a mass +/- a defect as shown below.

Well…now you will notice the little min/max check box on the mass search range. You can simply select the check box and it will switch to input the range from minimum to maximum as shown below.

To switch back just deselect.

This was a direct request from one of our users for usability. Glad to oblige.

Search Structures With Associated Patents Only

ChemSpiderMan — Fri, 09 Nov 2007 05:10:32 +0000

I have reported previously on the integration between the ChemSpider site and the SureChem online patent database (1,2,3). What is unique about the SureChem patent database is that it is searchable by structure and substructure as well as text. By indexing the structures from SureChem on ChemSpider we can allow search results on ChemSpider to link directly to the related patent on SureChem. One click away.

We have now added the ability to search ONLY structures with associated patents. In the Advanced Search Mode simply click on the appropriate option “Filter only those having patents link”. Enjoy!

Recent Downtime Events

ChemSpiderMan — Mon, 29 Oct 2007 11:36:44 +0000

Some of you might have been affected by ChemSpider service outages over the weekend. We are used to experiencing outages as a result of power cuts or internet service provider outages. Unfortunately this one is our of our making. There were numerous things going on over the weekend and we were hoping for only slight interruptions. We have had to add another terabyte of storage to the system and needed to pull down the system to put the drives in place. Also, we are in the process of rebuilding our index for the Chemrefer service and this is putting a strain on our servers. Add to this the fact that Google appear to be indexing all of our InChIKeys (we estimate about 10,000 transactions per day on the server just from this indexing process). As commented previously ChemSpider runs skinny and we don’t have an enormous rack of servers, we are limited to 3. We apologize for the outages but until we can afford to have ChemSpider run as a hosted service this may happen from time to time.

Systematic Names Available Again on ChemSpider

ChemSpiderMan — Mon, 29 Oct 2007 03:59:36 +0000

When ChemSpider went live earlier this year all chemical structures had systematic names generated. These were later removed. We are happy to announce that OpenEye have been kind enough to provide us with a copy of their Lexichem. This software will allow us to generate systematic names for structures across the database. The structures are presently being named in batches and you will see them show up as we publish them to the database. An example systematic name is shown below.

Common Names are now available on ChemSpider

ChemSpiderMan — Mon, 29 Oct 2007 03:05:47 +0000

There are literally millions of chemical names, identifiers and registry numbers associated with the chemicals on ChemSpider.As I have blogged many times before there are efforts afoot to improve the quality of the data and this is actually going quite well. As part of our effort to improve the data we are allowing one of the identifiers associated with each of the structures to be declared the common name. When this is done then the common name is highlighted by drawing a box around it and moving it to the TOP of the list. One more step to help chemists know what a chemical is…

Adding Identifiers to Existing Records

ChemSpiderMan — Mon, 29 Oct 2007 02:46:58 +0000

Over the past few months we have continued to add layers of curation and annotation to the records on the ChemSpider database. We have enabled direct data curation, the addition of spectral data and a beta version of new structure submission.

We have been spending a lot of time working on cleaning existing identifiers on the ChemSpider database. However, an obvious but lacking feature was the simple ability to add a new identifier to an existing record. So…it’s done now..

To add an identifier simply Login to the system and search on the structure of interest. if you know there is an identifier missing simply click on: Add Identifier. When you do so a box will open allowing you to add one or more identifiers (one per line please) as shown below.

Unless you have curating permissions you will not see the check box to approve the identifier. That is left to the curators. Simply fill in the captcha, the two word confirmation box in place to keep spammers at bay, and the identifiers will be added to the list. You have just contributed to the list of chemical names on ChemSpider. Thank you!

JSpecView Visualization Issues Now Resolved on ChemSpider

ChemSpiderMan — Wed, 10 Oct 2007 23:12:26 +0000

ChemSpider users are aware of the fact that we have supported the deposition of spectra to ChemSpider so that structure-spectra association is in place. We have described this in previous posts (1,2) and a technical note about depositing spectral data is available online.

Thanks to the valiant efforts of Bob Lancashire it appears that he has resolved the issues regarding Java versions 1.5 onwards. In order to see spectral viewing in place please visit the cocaine structure here and see the five associated spectra. Thanks Bob!

3D Optimization Now Embedded Underneath JMol on ChemSpider

ChemSpiderMan — Tue, 25 Sep 2007 01:27:33 +0000

A blog posting last night discussed the fact that we had not yet implemented a 3D optimization engine under Jmol and that the structures displayed are 2D FLAT molecules. Well I’m happy to report that less than 24 hours later (actually it was done before I woke up this morning) the optimization of 3D structures under JMol is incorporated.

Rajarshi Guha pointed us to smi23d and we have embedded this under JMol now. This is of course, beta (we stay in constant beta) and there are a couple of bugs…we’re not sure whether these are on our side or in the Optimizer and are awaiting feedback. In the previous posting on Diazonamide A I commented on the six hits. The top two do NOT optimize while the bottom 4 do. I think it’s about the fact that the structure might be TOO far away from a starting point for optimization. I’ve opened an email dialog with Rajarshi to discuss.

Please note that the 3D coordinates are calculated in real time and the JMol screen might stay empty for a while during the optimization process. Here’s what it looks like for Diazonamide A.