Comments on: The Entire ChemSpider Database is On Its Way to PubChem!

By: M Karthikeyan

M Karthikeyan — Tue, 11 Dec 2007 01:43:07 +0000

Last friday I did update the Pubchem data (FTP download of compounds) locally in a oracle database server using Jchem (chemaxon tools). my statistics (excluding the error structures deducted by Jchem) is reproduced below (total: 11,623,278) . Each row represent one million counts. Please note that 13-16 has low count of actual entries for every one million molecules expected. To my surprise today the total list of molecules available for download at Pubchem (compounds) increased to 23 millions. I am in the process of updating the local database for actual counts.

x10+6 Entries
1 863132
2 653990
3 755562
4 624029
5 961141
6 609013
7 601475
8 937742
9 985608
10 882349
11 998475
12 996339
13 11865
14 3499
15 2582
16 59521
17 920222
18 756734

any update on pubchem is welcome.
M.Karthikeyan

By: Unilever Centre for Molecular Informatics, Cambridge - petermr’s blog » Blog Archive » Chemspider and Pubchem - open data

Fri, 30 Nov 2007 09:30:45 +0000

[...] ChemSpider Blog » Blog Archive » The Entire ChemSpider Database is On Its Way to PubChem! [...]

By: Antony Williams

Antony Williams — Fri, 30 Nov 2007 03:08:19 +0000

Rich,

I believe PubChem update on a fairly regular basis but they’ve never dealt with over 17.5 M structures so I don’t know how ling it will take them.

Regarding the linkage to patents – PubChem will link back to ChemSpider via the ChemSpider ID and Data Source connection. Since the SureChem patent structures are on ChemSpider users will land from PubChem onto ChemSpider and only determine the link to patents at that point. The link from PubChem to SureChem patents directly would come if SureChem chose to deposit their own structure collection directly.

Regarding “This sets the bar very high for Open data in chemistry. I’m not sure what to call it, but it’s a game-changer.” You’ve likely seen over the past few months we’ve been challenged on Open Access and Open Data. We’re declaring neither. We’re declaring Free Access. Our Web APIs have opened up access to certain data only. If people choose to scrape other data such as predicted data they are violating rights of other groups who have allowed us to use their algorithms. While SureChem remains free access then mashups of some form might be possible but would likely take a lot of work.

Thanks for the compliments Rich..they are always welcome.

By: Rich Apodaca

Rich Apodaca — Thu, 29 Nov 2007 16:10:54 +0000

Antony,

This is great news. Any word on when the first structures will be appearing? How frequently will PubChem be updated with new ChemSpider content?

Also, through what mechanism will individual PubChem compounds be linked into patents and the primary literature?

For example, I’m familiar with PubChem Substance records. Two fields that allow URLS are:

PUBCHEM_EXT_DATASOURCE_URL
PUBCHEM_EXT_SUBSTANCE_URL

So one way would be for Substance records to link back to the appropriate ChemSpider compound summary page. Another would be for ChemSpider compound IDs to be added to the PUBCHEM_EXT_DATASOURCE_REGID field. Or will another mechanism be used?

Regardless of how exactly linkage occurs, the end result would be that any third party could, independently of ChemSpider, reconstruct the entire ChemSpider compound database. By using the ChemSpider Web APIs, they could develop a parallel service that re-processes the ChemSpider analytical data and patent/primary literature data, possibly mashing up the data from other sources as well.

This sets the bar very high for Open data in chemistry. I’m not sure what to call it, but it’s a game-changer.

Great work!