Chemical Inventory Systems: Cheminformatics' Problem Child?

June 08, 2010

Chemical inventory is one of those things that few in cheminformatics think about but many in chemistry do. Having worked in research labs in both industry and academia, it seems that chemical inventory, and particularly the software/hardware to manage it, is one of those universal problems in chemistry.

What Problem?

What problem, you may ask? The problem of finding a sample of a given substance present somewhere in a research lab. Perhaps you need a TLC or NMR standard. Maybe you're interested in starting a synthesis and don't want to waste money buying what you already have. You might even have just run out of your personal supply of a reagent and need to get more quick.

When a member of a lab buys a substance, often the sample is not completely used and so needs to get stored in common storage. Over time, a substantial inventory of chemicals can accumulate. This inventory can be a powerful tool for getting things done faster, cheaper and safer - but only with a good system for finding what you're looking for.

One of the new product ideas we're kicking around at Metamolecular is a simple, inexpensive, hosted, browser-based chemical inventory management system. So we're looking to see how big this problem might be and how well existing solutions do the job.

Where There Are Many Treatments, There Are Few Cures

It's not like solutions don't exist; just googling 'chemical inventory' turns up a number of companies in the search results (and through paid ads) with products they say work quite well, thank you very much.

And it's not just companies offering solutions, either. For example the ChemTracker consortium, a non-profit organization, offers a hosted system that apparently is being used by a number of universities in the US.

But having used a couple of these chemical inventory systems in the past, and having listened closely to what was being said about them by others, I'm struck by how badly - in general - they seem to do the job. Everything from outright bugs that go unfixed to poor vendor support to frightful usability problems are cited.

Practical Cheminformatics

This state of affairs seems surprising given that in any chemical tracking system you'll find a number of great cheminformatics problems to sink your teeth into: substructure search; structure similarity; structure rendering; structure to name conversion; and unique identifiers. (For still more on the subject, see the D-F series on the design process for a chemical tracking system) Then there's the fact that this kind of software, done well, can make a real difference. Once implemented, a chemical tracking system becomes an integral part of lab chemists' workflow.

What About You?

If you have views on chemical inventory software and would like to share them, we would like to hear from you by email or the web. If you work in a lab now, what chemical tracking system do you use? What do you like best about it and what would you change?