Milot Mirdita

Milot Mirdita

Postdoc in Bioinformatics

Seoul National University

I am a postdoctoral researcher at Seoul National University, where I work with Martin Steinegger. My expertise includes bioinformatics, metagenomics, protein structure prediction and machine learning. I’m a developer and maintainer of many open-source tools like ColabFold, MMseqs2 and Foldseek. I hold a PhD from the University of Göttingen and the MPI-NAT, where I worked with Johannes Söding.

Education

 
 
 
 
 
Postdoc
August 2022 – Present Seoul, Korea
 
 
 
 
 
Ph.D. (Dr. rer. nat, summa cum laude)
July 2017 – April 2022 Göttingen, Germany
 
 
 
 
 
M.Sc. in Computer Science
April 2014 – August 2016 Munich, Germany
 
 
 
 
 
B.Sc. in Bioinformatics
October 2010 – March 2014 Munich, Germany

Projects

*
ColabFold: making protein folding accessible to all
ColabFold accelerates protein structure and complex prediction by utilizing the fast homology search of MMseqs2 and the advanced folding model of AlphaFold2.
ColabFold: making protein folding accessible to all
Foldseek: fast and accurate protein structure search
Foldseek is a free open-source software and webserver that enables fast and sensitive comparisons of large sets of publicly available protein structures, with sensitivities similar to state-of-the-art structural aligners but four to five orders of magnitude faster.
Foldseek: fast and accurate protein structure search
Plass: Protein-level assembly increases protein sequence recovery from metagenomic samples manyfol
Plass is an open-source assembler that can recover 2-10 times more protein sequences from complex metagenomes, and has been used to assemble some of the largest free collections of protein sequences.
Plass: Protein-level assembly increases protein sequence recovery from metagenomic samples manyfol

Publications

(2025). De novo discovery of conserved gene clusters in microbial genomes with Spacedust. Nature Methods.

Cite DOI URL

(2025). GPU-accelerated homology search with MMseqs2. Nature Methods.

Cite DOI URL

(2025). AlphaFold Protein Structure Database and 3D-Beacons: new data and capabilities. Journal of Molecular Biology.

Cite DOI URL

(2025). Easy and accurate protein structure prediction using ColabFold. Nature Protocols.

Cite DOI URL

(2025). Easy and interactive taxonomic profiling with Metabuli App. bioRxiv.

Cite DOI URL

Contact