A professional tool for comprehensive statistical evaluation of MS and MS/MS data

Mass spectrometry is an essential analytical technique for high-throughput analysis in proteomics and metabolomics. Not being able to properly analyze the massive amounts of data provided by this technique is currently often the bottleneck for experimental studies. Although software tools for many data analysis tasks are available today, they are often hard to use, hard to combine with each other, and not flexible enough to allow for rapid analyses of large number of highly complex samples.

Medicwave Bioinformatics Suite™ (MBS™) combines a large set of data processing, data exploration and data mining techniques with a rich set of statistical evaluation methods and a comprehensive amount of methods for statistical presentation and visualization of results, all within an easy-to-use software framework.

Software tool for Quantitative Proteomics in MALDI MS / LC-MS profiling experiments

In collaboration with Professor Knut Reinert (from Free University Berlin) and Professor Oliver Kohlbacher (from Eberhard Karls University Tübingen) we have combined our Medicwave Bioinformatics Suite™ (MBS™) for multivariate statistics with OpenMS - a C++ library for LC-MS data management and analysis. OpenMS was implemented into MBS™ under the name LEQuant™.

Before Data Mining:

MBS™ is able to reads many different LC-MS data formats, from MzData, MzXML, XML, DTA and T2D, to certain RAW formats such as Thermo Xcalibur® RAW and Bruker® FID. Filters are used to reduce chemical and random noise as well as baseline trends. Algorithms are thereafter applied to detect peptidic features in LC-MS data, and then to align multiple experiments with each other as well as to match the corresponding ion species across many samples.

Data Mining:

Powerful pattern recognition techniques and comprehensive statistical evaluation methods are used to filter out the relevant information from different sample groups. Using different data mining tools e.g. dimension reduction methods such as PCA and PLS or classification methods such as LDA, statistically relevant features (biomarker candidates) can be detected.

The quantitative analysis of MS data often results in the discovery of interesting peptides (biomarker candidates) by differentiating between different groups e.g. patients samples and control samples.

After Data Mining:

Thereafter, MS data are linked with MS/MS data. The discovery of interesting peptides can be followed by database searches for protein identification and de novo sequencing using MS/MS data.

Pre-Processing and Data Mining