RSS

Category Archives: Consulting

Chemical Information Mining Book – A Moment of Pride

As I recall last Christmas I was finishing up a chapter in a book that is now on the market and called Chemical Information Mining. What’s amazing is that I just found that the first 21 pages are ALREADY on Google Books. I could’ve seen the cover there before waiting to receive a copy…hell I could’ve read UP to my Chapter…that’s where it stops.

The product description on Amazon is: “This book focuses on information extraction issues, highlights available solutions, and underscores the value of these solutions to academic and commercial scientists. After introducing the drivers behind chemical text mining, it discusses chemical semantics. The contributors describe the tools that identify and convert chemical names and images to structure-searchable information. They also explain natural language processing, name entity recognition concepts, and semantic web technologies. Following a section on current trends in the field, the book looks at where information mining approaches fit into the research needs within the life sciences.”

I’m rather proud of the contribution Andrey Yerin and I made to the book. I worked with Andrey while I was at ACD/Labs and learned all about nomenclature from him. He’s one of the nicest, most competent and focused specialists in the domain of systematic nomenclature in the world. The book chapter contents are listed below. Makes for good Xmas reading if you care about that type of thing…

 
1 Comment

Posted by on December 18, 2008 in Book Reviews, Consulting

 

Tags: , ,

Retrosynthetic Analysis Presentation at ACS-Philly

I had the pleasure of representing ARChem Route Designer, a retrosynthetic analysis tool from SimBioSys at the American Chemical Society meeting in Philadelphia last week. While I am not an organic synthetic chemist I have done my fair share of syntheses during my BSc and PhD and actually had a bit of a green thumb when it came to purity and yield. When given the opportunity I ran instead at exciting nuclear spins in large magnets (and enjoyed my choice for many years).

Since I was in the commercial sector for over a decade managing the development of chemistry software I have always had an interest in the development of a retrosynthetic analysis tool. It’s a lot of work and requires a deep understanding of organic chemistry. ARChem results from combining the deep understanding of Peter Johnson with the software development skills of SimBioSys. Peter was not able to make it to the ACS meeting in Philadelphia and, since I have had some experience of ARChem as a result of working with SimBioSys over the past few months, I was asked to step in and present on the product.

 

A link to the presentation is given here. A paper on ARChem has also been submitted if anyone is interested.

 
1 Comment

Posted by on August 24, 2008 in Computing, Consulting

 

Tags:

A Green Solution for Virtual Screening Using the IBM Cell Broadband Processor

I spent a few days this week in Boston at the Bio-IT conference. I was there for two reasons – to support one of the companies I have been consulting with of late and to present on ChemSpider.

The ChemSpider presentation seemed to be well received and I’m grateful for the opportunity to expose the ongoing work we are doing on ChemSpider.

I have spent the past few months supporting the efforts of SimBioSys to bring a potentially revolutionary platform to market. The intention was to deliver the platform to Boston Bio-It and under the leadership of Zsolt Zsoldos, the Chief Technology Officer for SimBioSys, the deadline was met. Zsolt’s already blogged about the WOW FACTOR at Bio-IT. Check it out…it was a true phenomenon.

I’ve blogged previously about the possibility to derive processing power from a gaming system (1,2). By the time that eHITS Lightning was unveiled at Bio-IT the Cell Processor had managed to deliver up to 120X performance improvements for certain examples.A White Paper about the technology has been online for a while now. The technology delivered was enough to garner a position as a finalist for the Best in Show award. It was enough to have some of the other domain players question how much work it was to port eHITS to the Cell. It was essentially a full rewrite and over two man years of effort to deal with coding for the special nature of the processor.

Zsolt’s presentation to the Best in Show judges is online here. You MUST look at slides 8 and 9 to really “get it”. The cost savings associated with the electricity, cooling, space and, in theory, network administration, are enormous. A 100 CPU Intel cluster could be replaced with three PlayStations for the same eHITS throughput. It’s definitely a  “Green Solution”.

With so much attention being given to coding on Graphic Processing Units it’s quite surprising that no one is yet talking about the advantages of the Cell processor. Well, maybe they will now!

 
1 Comment

Posted by on May 2, 2008 in Consulting

 

Tags: , , ,

 
Stop SOPA