Bioinformatics Laboratory Medical Bioinformatics and e-Bioscience

Department of Clinical Epidemiology, Biostatistics and Bioinformatics

Academic Medical Center

University of Amsterdam

Go to Blackboard

If you are interested then contact Antoine van Kampen (a.h.vankampen@amc.uva.nl) to discuss interests and possibilities.

See also Research projects for master students (Biological, Biomedical, and life sciences)

• Bioinformatics analysis of LC-MS metabolomics data
  • Supervisor: Antoine van Kampen, a.h.vankampen@amc.uva.nl, 020-5667096
  • Description: Liquid chromatography coupled to mass spectrometry (LC/MS) is used in metabolomics research. In this context, the technology has been increasingly used for e.g, the discovery of biomarkers. One of the challenges in this domain remains development of better approaches for the bioinformatics analysis of LC/MS data. In this project we aim to develop a processing pipeline for pre-processing and statistical analysis of metabolomics data.
  • Technical skills: The students should have programming skills and interest in bioinformatics. Knowledge and interest in statistical analysis is important. The methods will be developed in the R-statistical package and we will explore the use of Taverna for workflow management.

• An in-silico approach for the detection of contaminated tumor cell lines
  • Supervisor: Perry Moerland, p.d.moerland@amc.uva.nl, 020-5664660
  • Description: For decades, hundreds of different human tumor type–specific cell lines have been used in experimental cancer research as models for their respective tumors. However, sometimes cell lines that have been used for years turn out to be contaminated (see here for a recent example). In this project you will work on an approach that uses publicly available microarray datasets of human cancer cell lines, primary tumors, and tissues to detect contaminated cell lines. The approach is based on the intuitive idea that often contamination can be traced back to different tumor types or tissues. We expect that these signatures of contamination can also be found in the cell line's gene expression profile. Our lab already has most tools in place to integrate multiple microarray datasets in a single database. Focus of this project would be on developing the methods for comparing cell line gene expression data with other datasets. The ultimate goal would be a tool to which a researcher can submit his cell line gene expression data and that scores the likelihood of the cell line being contaminated.
  • Technical skills: knowledge of bionformatics in general and specifically the statistical programming language R would be an advantage. Following our MSc level course at the University of Amsterdam on the analysis of genome-wide experiments provides most of the required skills.