Photo courtesy of Flickr user Horia Varlan.
Since partnering with TIBCO Software Inc. in 2012 to bring data visualization and analysis to scientific industries, the PerkinElmer Informatics team has been working to develop add-ons and methodologies to accelerate scientific research using the TIBCO Spotfire® software platform.
Our team has devised a new way of using the TIBCO Spotfire software platform to dynamically mine reaction data contained within ELN. When reaction data is entered into an ELN, the stoichiometry is calculated and linked to an autotext feature that automatically updates as reagent or product amounts are changed. Solvent, temperature, pressure, atom economy and conditional information are also captured and stored in ELN. But with all of the complex reaction data contained in an ELN system, traditional reaction searching methods can omit relevant information in search results.
In a typical query form, scientists describe a reaction and its parameters which generates a list of search results, ranked in the order in which the system calculates its relevance. The results show experiment name, the reaction, and other metadata for each entry, and by clicking on an entry, a user can go to it and view it in detail.
But using this traditional search method, scientists are dependent on the system’s “best-result” ranking system. The search list doesn’t update when criteria - such as percentage yield – is changed, meaning to modify results, the entire search must be re-run. Furthermore, this list reflects the type of information inherent to an ELN by generating a list of experiments, when reaction information might be more useful when populated in a table of structured data. Finally, there may be data relevant to a reaction search that is housed in other sources, such as an Inventory, Registration, or another ELN. Traditional search methods simply can’t access and compile these various sets of data.
By integrating the TIBCO Spotfire software platform with ELN, reaction searches use dynamic filters to yield interactive visualizations of data presented in hierarchical data tables. Results can be drawn from multiple databases and merged using Information Links or Datalytix. The TIBCO Spotfire software platform is already widely used across chemical and pharmaceutical industries, but it has not yet been used to dynamically mine ELN-based reaction data in this novel manner.
The PerkinElmer Informatics team has tested this reaction-searching method using sample datasets and has proven that TIBCO Spotfire successfully achieves improved search results. Our team is now seeking a partner interested in being an early adopter of this methodology to implement the first real-use case of dynamic reaction searching in ELN using data visualization software.
If you are interested in learning more about whether your laboratory would make a good candidate for the first implementation of this methodology, contact us for more information here.