Project details


Method Validation for Time Alignment Algorithms

proteomics bioinformatics statistics time alignment mass spectrometry

C. Christin
prof. dr. R.P.H. Bischoff

Type of project:
Stage Wetenschap / Research project

Nature of the research:
This is a bioinfromatics project requiring skills in Matlab and Java programming, (bio)statistics and mathematics.

Fields of study:
laboratory medicine general internal medicine biomaterials

Background / introduction
Time alignment is a part of data preprocessing that corrects for retention time shifts between chromatograms from different runs in a dataset. The development of proper time alignment algorithms is compulsory in LC-MS based biomarker studies. Several modification of existing non-linear time alignment methods have been implemented in our laboratory. However, the continuous improvement of time alignment algorithms is necessary in order to obtain better quality processed data, specially for datasets with a high variation of compound concentrations. For this reason we are currently modifying several time alignment algorithms using one-dimensional information for the benefit function to obtain the optimal time shift correction path. The modifications involve the use of two-dimensional benefit functions separating signals of closely eluting compounds but having different mass.
Research question / problem definition
The main work of a student in this project will be to assess the quality of currently developed time alignments using complex datasets that have been obtained from ongoing biomarker research. Secondly the student will be involved in the development of visualization tools to compare and assess local and global time alignment quality. It is expected that the work of the student will become part of a publication.
The student will work directly with Christin Christin performing PhD research in the domain with close supervision of Peter Horvatovich.
back to toptop