Philpp C. Münch

Research

My current work focuses on advancing deep learning for genomic applications. In collaboration with colleagues, I introduced Self-GenomeNet (a self-supervised approach that leverages reverse complement sequences), GenomeNet-Architect (an automated neural network optimization platform), and deepG (an R library to facilitate data handling and model selection in genomics). Recently, I’ve explored using large language models to expand phenotype annotations for microbial species—sometimes outperforming purely sequence-based strategies—while making these resources available via a shared portal for the research community.

Personal Interests

I live in Munich, Bavaria, where weekend hikes in the Alps are a welcome break from my work. I share my day-to-day life with a Bernese mountain dog, who keeps me moving and helps me appreciate the balance between research and the outdoors.

I enjoy reading Astral Codex Ten and occasionally browse LessWrong for new perspectives on science and related topics. I also like listening to Vivaldi, admiring Renoir’s Impressionist art.

If you’d like to discuss potential collaborations or just say hello, feel free to email me at muench@hsph.harvard.edu. I’m always happy to connect with fellow researchers and enthusiasts.

Selected publications

• Gündüz, H.A., Binder, M., To, X.Y., et al. (2023). A self-supervised deep learning method for data-efficient training in genomics. Commun Biol 6, 928. [doi]
• Münch, P.C., Stecher, B., McHardy, A.C., et al. (2021). Identification of Natural CRISPR Systems and Targets in the Human Microbiome. Cell Host & Microbe 29, 94–106.e4.
• Münch, P.C., Eberl, C., Woelfel, S., et al. (2023). Pulsed antibiotic treatments of gnotobiotic mice manifest in complex bacterial community dynamics. Cell Host & Microbe 31, 1007–1020.e4.

More details and a complete list: Google Scholar