Use the NPC browser content as your Walmart shopping list?

This blog feels like GroundHog day. The last one on the topic I hope.

A long time ago (or so it feels, actually a few months) the NPC browser was released. A monumental effort and achievement that took years of work, dedication and your tax dollars. Almost immediately, Antony Williams (ChemConnector) , a scientist who has spent many hours cleaning up chemistry structures on the web (and basically improving the quality of what is out there), found some very significant errors in it within seconds of looking at the structures. These problems could have been found by anyone skilled in the art of cheminformatics. We both then blogged and highlighted this lack of curation to alert the community as fast as possible..and took some flak in the process. We also tried to publish these observations in the same journal [science translational Medicine] NCGC did, but were rejected – (more on that some other time). It took a month or so for NCGC to add a disclaimer to their website. They then removed the disclaimer and released a new version (hooray) and thanked the few people who helped them do what they should have done in the beginning. It all went quiet for a while and then recently Francis Collins mentions this effort for reengineering translational science. All was well, rosy and curated in NPC browser world. Or so you are lead to believe.

Jump to today. Antony has been continuing to look at the browser and found some odd content. He was just skimming the surface finding VEGETABLE listed..I downloaded version 1.1.0 and found…. radish, Sea minerals, vitamins, chinese rhubarb, neon, Prones-pasta aroma, Aqua marina, egg yolk, corn syrup, egg white, english walnut…should I go on. The stake through the heart was cockroach, American followed by cotton swab, beet and crabs! Is it a shopping list for Walmart! And of course none of these have molecule structures (so they definitely got that right). OK so we are being picky – definitely..but is anyone at the NIH reviewing this database before it goes out. Anyone skilled in the art of reading could apply for that job, no Ph.D required.

Should these objects be in there to begin with- what use are they – why oh why? Please clean up this database NCGC and I will stop blogging about it and giving you more free publicity!

No comment yet

2 pings

  1. Antony Williams, ChemConnector says:

    You are a funny guy. I can now honestly say that while I talked about a rabbit on my blog post you talked about the crabs you found one afternoon. Hilarious.

    I admit that I am no longer surprised by what I am finding in the database but I scratch my head as to how the results do not align with the paper in Science in Translational Medicine. The paper as written represents what seems like a good approach to assembling the data with a few caveats. But how did the data get to be this bad? If you ever get sick come visit me….based on what are listed as drugs I’ll be able to cure you by sitting you next to a rabbit, with crabs, eating vegetables, washed down with some beet juice sucked off a cotton swab. Yumm…

  2. sean says:

    A Comprehensive Resource of Clinically Approved Drugs Enabling Repurposing and Chemical Genomics
    does not =
    A Comprehensive List of Plants, Fruits, Vegetables and other ephemera

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>