I am Bioinformatician, intrigued by machine learning, algorithms, and software development.
In the age of big data, we are flooded with more data than we are able to comprehend. One major data source is biology, which has been digitized by high-throughput sequencing technologies for rapidly sequencing the whole genome of all creatures. The exponentially growing amount of biological data poses complex computational and scientific questions, which I aim to address by modern computer science and machine learning.
Currently, I am pursuing a PhD degree in Bioinformatics at the University of Cambridge. My supervisor is Oliver Stegle, and my co-supervisor is Zoubin Ghahramani. My research topics are machine learning models for analysing heterogeneous biological data, such as single-cell genomics, transcriptomics, and epigenomics. Specifically, I am interested in unsupervised methods for learning latent features from high-dimensional data.
Before, I studied Bioinformatics at the Technical University Munich (TUM), and Ludwig Maximilian University (LMU) Munich. My Master’s thesis was supervised by Fabian Theis at the Helmholtz-Zentrum Munich, which was about supervised machine learning for predicting the risk of type 1 diabetes from high-throughput data. My Bachelor’s thesis was supervised by Johannes Soeding at the Gene Center Munich, which was about conditional random fields for improving protein sequence searching.
During my studies, I worked as a research assistant in the Soeding group for computational biology. There, I maintained the Bioinformatics Toolkit, implemented HHfuncs for predicting functional sites in proteins, and extended CS-BLAST by a new model for predicting protein mutation probabilities.
