Overview:
The proposed master's thesis aims to explore the application of Timber Gridshells [1], focusing on the behavior of wooden elements subjected to a combination of torsional and tensile loading, as shown in (Figure 1). The study encompasses both experimental and numerical simulation aspects to validate its findings comprehensively. The experimental phase consists of two main parts: testing ash boards with finger joints (a subset already tested, as illustrated in Figure 2) and testing clear ash boards without finger joints with a specific focus on the impact of knots. Experimental results will undergo validation through Finite Element Modeling (FEM). A preliminary FEM model shown in (Figure 2) has been established to guide the modeling process, particularly concerning tension behavior with torsional effects in the case of finger joints [2]. However, this model requires further refinement and thorough validation against experimental data to ensure accuracy and reliability.

Tasks:

•  Literature Review: Review existing research on Timber Gridshells, focusing on combined torsional and tensile loading conditions. Reviewing the previous research on modeling finger joints and their interaction properties.
• Experimental Testing: Complete testing on ash boards with finger joints. In addition, conduct experiments on clear ash boards without finger joints to study knot impact under the same loading condition.
• Numerical Simulation: Develop Abaqus models to simulate experimental scenarios and validate numerical results with the experimental data for accuracy.
• Documentation and Analysis: Document experimental procedures, results, and numerical simulations. Analyze the findings, establishing correlations between experimental and numerical data.
   

  Apply now and submit your CV and grade report to Mostafa Abdelrahman (mostafa.abdelrahman(at)tum.de) and Ani Khaloian (sarnaghi(at)hfm.tum.de).
   

  [1]   B. D’Amico, A. Kermani, and H. Zhang, “Form finding and structural analysis of actively bent timber grid shells,” Eng. Struct., vol. 81, pp. 195–207, Dec. 2014, doi: 10.1016/j.engstruct.2014.09.043.

  [2]    V.-D. Tran, M. Oudjene, and P.-J. Méausoone, “FE analysis and geometrical optimization of timber beech finger-joint under bending test,” Int. J. Adhes. Adhes., vol. 52, pp. 40–47, Jul. 2014, doi: 10.1016/j.ijadhadh.2014.03.007.

  [3]    P. J. Cruz, Ed., “The nature of tectonic architecture and structural design,” in Structures and Architecture, 0 ed., CRC Press, 2013, pp. 266–273. doi: 10.1201/b15267-32.

  [4]    E. Schling, D. Hitrec, and R. Barthel, “Designing Grid Structures Using Asymptotic Curve Networks,” in Humanizing Digital Reality, K. De Rycke, C. Gengnagel, O. Baverel, J. Burry, C. Mueller, M. M. Nguyen, P. Rahm, and M. R. Thomsen, Eds., Singapore: Springer Singapore, 2018, pp. 125–140. doi: 10.1007/978-981-10-6611-5_12.

Overview:
Wood is a widely used construction material due to its renewable nature and favorable mechanical properties. However, its thermal behavior under high temperatures, especially during fire incidents, remains a critical aspect that requires comprehensive investigation [1, 2]. This master thesis aims to study the temperature-dependent material properties of wood under elevated temperatures. By subjecting wood samples to various temperature ranges, a deeper understanding can be obtained about moisture evaporation rate in the samples and their density loss. This thesis will contain two main parts, including experimental fire tests and CT-scanning of the samples . From the CT-scanned samples, models will be developed to link the dry density of the material to the stiffness properties. In addition, algorithms will be developed to distinguish between the dry and the wet density of the wood. Figure 1 depicts a previous study involving experimental testing, analysis, and simulation of the charring rate of Laminated Veneer Lumber (LVL) panels [3].

Tasks:

• Conduct a comprehensive literature review to gather existing knowledge on the thermal properties of wood and its behavior under different temperature ranges.
• Understand the CT scanning concept and the possibilities to distinguish between the dry and the wet densities.
• Design and set up controlled experiments to subject wood samples to different temperature ranges.
• Perform CT scans on the wood samples after different temperature ranges to visualize internal material changes
• Analyze the CT scan data to quantify density loss and density variation within the wood samples due to thermal exposure.
• Come up with an approach to distinguish between the dry and the wet wood densities from the CT scans, and to assign correct material properties based on the density values.

We offer:

• Thesis in the area that is highly demanded for further industrial development.
• Equipped workspace during the thesis period.
• Supervision and support during the thesis.
• Access to different systems required for the work.
• Opportunity to publish a research paper.

Apply now and submit your CV and grade report to Mostafa Abdelrahman (mostafa.abdelrahman@tum.de) and Ani Khaloian (sarnaghi@hfm.tum.de).

_{[1] United States Department of Agriculture and Forest Service, Wood Handbook, Wood as Engineered Material. USA: Forest Products Laboratory, 2010.
[2] A. Frangi and M. Fontana, “Charring rates and temperature profiles of wood sections,” Fire Mater., vol. 27, no. 2, pp. 91–102, Mar. 2003, doi:
10.1002/fam.819.
[3] M. Abdelrahman, Master Thesis, Fire performance of wood steel hybrid elements: Laboratory experiments and numerical simulation. TUM School of
Engineering and Design, group of wood technology, 2022.}

We (Benz lab, Gagneur lab) are looking for a motivated student of biology/molecular biotechnology/bioprocess engineering/biochemistry to assist in developing and carrying out a large, multi-species investigation of regulatory genomics.

The Project

Deciphering the impact of variation in non-coding regulatory regions of the genome on gene expression is a central challenge in genetics. This goal has not been achieved for any organism but important progress has been made by leveraging genome-wide omics datasets (RNA-Seq, ChIP-Seq, ATAC-Seq and quantitative proteomics, see e.g. [1]) and artificial intelligence, particularly deep learning ([2–4], see [5] for a review). A key challenge, however, is the limited amount of available sequence data with high variation, as most omics experiments are done in the same few model organisms. As such, deep learning models are only aware of a tiny fraction of the vast space of possible gene regulatory sequences.

We propose to move beyond these limitations by collecting genome-wide multi-omics data (RNA-Seq, ATAC-Seq and Mass-spectrometry based proteomics) from diverse fungal species in a variety of growth conditions. Based on these experiments, we will train a deep learning model to predict these omics observations from sequence. By jointly training on many species, our model can leverage both conservation and variation between species, to better learn the effect of individual regulatory elements on gene expression. We will use existing data, which leverages expression differences between closely related yeast strains and massively parallel reporter assays, to evaluate the model’s ability to predict the impact of sequence changes on expression [6,7]. Our ultimate goal is to crack the regulatory code of simple eukaryotes.

Your Tasks

• Growing cultures of diverse yeasts and filamentous fungi in the Benz lab
• Preparation of samples for NGS (RNA/ATAC-Seq) and Mass-spectrometry assays

• Depending on your interest and prior programming experience, contributing to the bioinformatics processing of the data, as well as the development, implementation and evaluation of the deep learning model (Gagneur lab)

Requirements

• Currently studying biology, biochemistry or a related subject at the MSc level
• Strong interest for regulatory genomics

• Prior experience with bioinformatics methods and machine learning is helpful, but not required

For any questions, please do not hesitate to contact us:
benz(at)hfm.tum.de (Benz) jobs-gagneurlab(at)in.tum.de (Gagneur)

1. ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489: 57–74.
2. Zhou J, Troyanskaya OG. Predicting effects of noncoding variants with deep learning-based sequence model. Nat Methods. 2015;12: 931–934.
3. Avsec Ž, Weilert M, Shrikumar A, Krueger S, Alexandari A, Dalal K, et al. Base-resolution models of transcription-factor binding reveal soft motif syntax. Nat Genet. 2021;53: 354–366.
4. Avsec Ž, Agarwal V, Visentin D, Ledsam JR, Grabska-Barwinska A, Taylor KR, et al. Effective gene expression prediction from sequence by integrating long-range interactions. Nat Methods. 2021;18: 1196–1203.
5. Eraslan G, Avsec Ž, Gagneur J, Theis FJ. Deep learning: new computational modelling techniques for genomics. Nat Rev Genet. 2019;20: 389–403.
6. Renganaath K, Cheung R, Day L, Kosuri S, Kruglyak L, Albert FW. Systematic identification of -regulatory variants that cause gene expression differences in a yeast cross. Elife. 2020;9. doi:10.7554/eLife.62669
7. Shih C-H, Fay J. Cis-regulatory variants affect gene expression dynamics in yeast. Elife. 2021;10. doi:10.7554/eLife.68469