Tony Williams grew up in Wales, UK - and originally considered being a Welsh teacher. However, after speaking with his career master, Tony quickly decided that career path probably wouldn’t lead to a successful career trajectory. Tony went to Liverpool University where he found Honors Level Chemistry fairly boring and spectroscopy especially challenging. He essentially self-taught himself spectroscopy from books and papers and, as fate would have it, Tony’s protons aligned to find his life’s passion in EPR/NMR spectroscopy. He later saw applications for this technology by studying Vitamin E related systems and this sparked a life long interest in NMR spectroscopy. He went on to get his PhD in NMR applying the technique to study lubricant systems at high pressures up to 5kbar.
Tony’s interest in computers and computing started when he was a poor graduate student. He wrote programs while getting his PhD. to analyze NMR relaxation data and installed PC’s in offices around central London. Tony spent some time at The National Research Centre laboratories in Ottawa performing Single Crystal EPR Rochester and finally landed in New York with the Eastman Kodak company. While at Kodak, Tony was the NMR Technology Leader for the company working with a team of almost 10 NMR scientists and almost as many NMR instruments. While at Kodak, he published over 20 papers and received 2 patents. One of his primary challenges was to homogenize instruments, platforms and processes at the company, specifically for walk-up spectroscopy laboratories so they could maximize throughput and ROI for the chemists. Tony was also one of the three people involved with building WIMS, a web-based information management system (http://www1.elsevier.com/homepage/saa/trac/wimsarti.htm) for managing samples, structure and spectra. This system still prevails today.
While at Kodak, Tony formed a close collaborative relationship with Advanced Chemistry Development (ACD/Labs) on a project to develop NMR based software. After a number of extremely fruitful interactions, Tony decided to take his entrepreneurial spirit out of a Fortume 500 company and into the small start-up company and joined ACD (www.acdlabs.com) as their NMR Product Manager. While at ACD/Labs, Tony managed the majority of their product lines with a focus on structure drawing and databasing, systematic nomenclature and analytical spectroscopy and chromatography. During his tenure he held multiple roles and was the Vice President and Chief Science Officer for a number of years. After a decade at ACD/labs and following the adoption of twin boys a couple of years earlier, Tony found that his demanding travel and work schedule didn’t fit the demands of his young family and started to look for other opportunities.
After seeing the efforts of both PubChem and Wikipedia to deliver chemistry-related data to the public his creative nature determined that maybe a combined approach of both efforts could deliver value to chemists worldwide. His initial concept was for a commercial software product that would crawl a network of disks internal to an organization and index chemistry in documents, reports and other files and integrate with data available on the web. However, Tony believes that a free-access website for indexing chemistry would have more value to Chemists. Thus, ChemSpider was born.
Here are excerpts from the interview:
Why did you found ChemZoo and produce ChemSpider?
[My] intention was always to form a community around chemical structures. We’d like to allow users to annotate and tag information around chemical structures to help share their knowledge and reduce the barriers, both cost, time and navigation hurdles, to chemistry related data. Presently we integrate to well over a 100 chemistry data sources with a few tens waiting to be added at present.
Can you give me some statistics on usage right now?
Presently we’re getting around 6000 users per day, and over 20,000 search requests a day (excluding robots). We’ve got over 1100 registered users but no one has to register to use the site. Those registered users however can add comments, deposit structures, spectra and so on.
How is the company funded?
We’re bootstrapping with personal bank accounts. Also, we have some sponsors who have been kind enough to contribute some donations, generally in exchange for us developing a particular aspect of the site such as extending the API.
Do you have a revenue model?
We’re figuring that out right now and have a number of different discussions with companies right now. This project started as a hobby project as I was gainfully employed at the time it started and never saw it as a source of income. Having chosen the path of self-employment as a consultant the potential of ChemSpider generating revenue is of course of interest. My preference is to not charge the users for public access but to obtain grants, maybe advertising revenue and require fees from companies who with to have secure links to the site for mashing results from ChemSpider data queries with their own internal systems. A rich uncle isn’t an option…I’ve looked across my family tree..
How many people work on the site? Are they full time?
[There are] 5 volunteers working on the site - they are involved with the coding, development of the website, algorithms for structure handling, text mining and so on. Then there are the users who help with “crowd sourcing” of data curation and deposition of new data
What do you want the site to be at the end of the day?
I’d like ChemSpider to be the first place people think of for researching information about small molecules. I work with the Wikipedia chemistry team, and I see our efforts to be complementary. We offer more capabilities however in terms of structure-based searching, hosting of analytical data, services for generation of properties, web-services for companies to integrate ChemSpider to their instruments and into their websites. It would be nice to be “ChemSPider Inside” to coin a phrase from Intel. Based on the interest of groups talking to us now you’ll see growth in the number of chemical vendors putting their catalogs online so it can become a public chemical directory for ourcing chemicals. We’re also working on making open access chemistry articles searchable by chemical structure. We’ve already done work with the International Union of Crystallography to make their open access article available via chemical structure searching. Chemspider is already a content system to aid Open Notebook science with thought-leaders like JC Bradley from Drexel university. We’ve just released the ChemSpider forum to allow the chemistry community to exchange information and collaborate. People struggling with synthesis or class of structures can get into collaborations. We are not short of ideas..bottom line though is that ChemSpider is intended to be one of the central communities to enable collaboration around chemistry and chemical structures
How many compounds are in chemspider?
We have 20 million compounds and growing literally daily.
Highlights?
My career highlights are many. I worked with excellent technical people at ACD/Labs and formed a rich network of friends in the community. Working with Dr Gary Martin from Schering Plough to crack a structure elucidation problem that remained unsolved for over a decade was great and the completion of a review article on Computer assisted structure elucidation that took over 18 months to write was just great.
Who should reach out to you in SciLink?
I pay my bills with consultancy so anyone wanting expertise in cheminformatics, systematic nomenclature and applications of software tools to chemistry is welcomed to contact me. In terms of ChemSpider we are always looking to support our users, extend and improve the system and collaborate with people interested in what we’re doing. I’m especially interested in talking to publishers who might want to deposit to ChemSpider and people who might be interested inus hosting their Chemistry information.
Comments(0) 
