The candidate will be responsible for the development of bioinformatics software to improve the comparability and reproducibility of state of the art methods to evaluate efficient data storage. That is to say, encoding information in biological molecules or chemical compounds. The action of encoding and decoding information is a process that we perform on a daily basis. Language is a great example as words carry different meanings. There exists many levels of processing information, the successful candidate will examine information theory metrics (e.g., compression ratio), error correction, and decoding accuracy (e.g., parity checks) for a quantitative assessment of information processing. A foundational aspect of this project is the creation of an evaluation framework that integrates state of the art methods for text mining (e.g., Markov chains, Long short-term memory or LSTM, etc). Her/His efforts include but are not limited to developing novel algorithms, implementing different state of the art methods into an open source framework, and maintaining a code base to benefit both the community and the research project.
In the research project MOSLA, the University of Marburg and the University of Giessen will jointly develop novel approaches and solutions for long-time archives based on molecular and chemical stor-age systems. Besides the technical solutions of data storage, they will also research in (de-)coding of information for long-time storage, which will be achieved by a combination of genetic and chemical information encoding. The project is funded by the Hessian Ministry for Science and Arts.