New image for “big data” tools

Big data represents both an opportunity and a challenge for scientists. The huge amount of data that can be generated by the latest scientific instruments could lead to important scientific discoveries, but only if scientists can efficiently process that data. This is especially true for analytical scientists studying biological materials using a fairly new technique known as mass spectrometry imaging (MSI). Help is now at hand, however, from a recent European project called Computis, which has come up with a range of new tools for efficiently processing MSI data.

As its name suggests, MSI is a form of mass spectrometry (MS), which enables both spatial and mass spectrometric analysis by measuring the masses of compounds at specific locations on the surface of a solid sample. MSI can thus build up a visual image of its chemical composition. Various MSI methods are available, including matrix-assisted laser desorption/ionisation (MALDI), secondary ion MS (SIMS) and desorption electrospray ionisation (DESI), but they all generate lots of data.

Although software tools are available for processing this data, they tend to be produced by the manufacturers of the MSI instruments and thus each uses different data formats, making it difficult to combine and process data produced by different instruments. Computis was set up by the European Commission in 2006 to overcome this problem by developing a new generation of flexible and efficient tools for processing any MSI data. It involved academic and industrial teams from across Europe, including the French Atomic Energy Commission and the Swiss pharmaceutical giant Novartis. The project finished in 2010 and the main outputs have now been unveiled in a new paper in EJMS—European Journal of Mass Spectrometry.

The project members quickly realised that their first task should be to develop a common format for MSI data that all the other data formats could be converted into, which led to the imzML data format. This format divides MSI data into two separate files: the mass data are stored in a binary file to ensure efficient storage, while metadata such as instrumental parameters and sample details are stored in an XML file. The members also developed a couple of tools for converting MSI data in other formats into the imzML format.

Next, the project developed several tools specially designed to work with the imzML format. This included two tools for processing and displaying MSI data, Data Cube Explorer and SpectViewer, and a tool called EasyReg2D for combining MSI data with image data from other analytical instruments such as microscopes. In addition, they adapted an existing MSI processing tool known as BioMap to work with data in the imzML format.

All in all, this project has succeeded in greatly expanding the tools available for processing MSI data, helping to ensure big data becomes an opportunity for scientists rather than remaining a challenge.

The research is published as:

M.-F. Robbe, J.-P. Both, B. Prideaux, I. Klinkert, V. Picaud, T. Schramm, A. Hester, V. Guevara, M. Stoeckli, A. Roempp, R.M.A. Heeren, B. Spengler and S. Haan, “Software tools of the Computis European project to process mass spectrometry images”, Eur. J. Mass Spectrom. 20(5), 351–360 (2014). doi: