Open Projects at the Bachelor, Master, Ph.D., and post doc level
We are always looking for talented and motivated students with backgrounds in physics, chemistry, biology, computer science and related disciplines who are interested in joining our lab for a range of projects. If you are interested, please contact Jan Lipfert with a CV, transcript, and short statement of motivation.
Highly parallized magnetic tweezers for single-molecule force spectroscopy
Master thesis project involving instrumentation, single-molecule measurements, and data analysis
Magnetic tweezers are a powerful tool to apply forces and torques to single molecules tethered between a surface and small magnetic beads. A particular strength of magnetic tweezers is that we can apply constant forces to and track many molecules at the same time. Aim of this project is to develop a massively parallel magnetic tweezers set up that can probe 1000-10,000 molecules at the same time, with single molecule resolution. This new instrument will give us the possibility to probe ligand-receptor interactions and protein folding under forces at an unprecedented scale.
Synthesis of monodisperse cubic-shaped magnetic nanoparticles for biophysical applications
Master thesis project at the intersection of nanoparticle synthesis and biophysics
Synthesis of uniformly sized magnetic nanoparticles for biological applications have taken a huge step in the past few years thanks to high temperature colloidal syntheses. In order to apply magnetic force and torques to molecules and cells, large magnetic nanoparticles (ca. 40-80 nm; see the figure for a TEM image) being able of exerting forces in the pN regime are required. The focus of this thesis lies in the colloidal synthesis of iron oxide nanoparticles with sizes ranging from 40 to 80 nm. We aim to understand the effect of ligands, solvent, and impurities on shape, size, and magnetization of the particles. The particles will be characterized with a variety of physico-chemical techniques such as transmission electron microscopy, dynamic light scattering, Fourier-transform infrared spectroscopy, X-ray diffraction and scattering.
Deep Learning DNA
Master thesis project at the intersection of machine learning and biophysics
Atomic force microscopy (AFM) imaging is a very-high resolution type of scanning probe microscopy that can image biological macromolecules with a resolution of ~1 nm. AFM imaging has been used extensively to visualize DNA (an example is shown in the figure above), for example to probe its mechanical properties and to visualize its conformations. In addition, AFM imaging has proven to be powerful in revealing DNA-protein complexes involved a large range of biological process (see Vanderlinden et al., 2014 for a recent example).
So far, the AFM images of DNA and nucleo-protein complexes have mostly been analyzed manually, by clicking on or tracing structures in the images. Aim of this master thesis project is to developed algorithms that can process images automatically, by extracting and scoring relevant features based on machine learning techniques. Such automated analysis has the potential to 1) dramatically increase the throughput of AFM imaging analysis by replacing time consuming manual inspection and 2) make the image analysis more reliable and statistically sound, by avoiding human judgment and biases.
A first application of the automated image analysis will be a full quantification of the topology of DNA plasmids (circular pieces of DNA present in bacteria; a typical image is shown above) to answer questions regarding the partitioning between DNA twist and writhe in these molecules (see Brouns et al. 2018 for a recent example). In a first step, we will transfer machine learning approaches based on convolutional neuronal networks that are used for vessel detection in medical image processing. If successful, the algorithmic approach can be extended to many other quantification tasks as well, for example to the analysis of the DNA-protein complexes or complex protein structures (see Müller et al., 2016 for an example).
This project is a collaboration between the labs of Prof. Björn Menze (Image-based biomedical modeling, TU Munich) and Prof. Jan Lipfert (Molecular biophysics, LMU Munich). We are looking for a highly motivated student from a background in physics, chemistry, or computer science who is interested in very interdisciplinary work and has basic programming experience in CPP and/or Matlab and/or Python. If you are interested, please contact us at Bjoern.Menze@tum.de or Jan.Lipfert@lmu.de, including a brief CV and transcript.
- Brouns T., De Keersmaecker H., Konrad S.F., Kodera N., Ando T., Lipfert J., De Feyter S., Vanderlinden W. Free Energy Landscape and Dynamics of Supercoiled DNA by High-Speed Atomic Force Microscopy. ACS Nano. 2018 Oct 29. doi: 10.1021/acsnano.8b06994
- Müller, J.P., Mielke, S., Lof, A., Obser, T., Beer, C., Bruetzel, L.K., Pippig, D.A., Vanderlinden, W., Lipfert, J., Schneppenheim, R., et al. (2016). Force sensing by the vascular protein von Willebrand factor is tuned by a strong intermonomer interaction. Proc Natl Acad Sci U S A 113, 1208-1213.
- Vanderlinden, W., Lipfert, J., Demeulemeester, J., Debyser, Z., and De Feyter, S. (2014). Structure, mechanics, and binding mode heterogeneity of LEDGF/p75-DNA nucleoprotein complexes revealed by scanning force microscopy. Nanoscale 6, 4611-4619.