Eric Kummelstedt
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Eric Kummelstedt has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Daniel Álvarez
Department of Biology, University of Fribourg, Chemin du Musée 10, 1700 Fribourg, Switzerland
1 shared publication - David Gfeller
Department of Oncology, Ludwig Institute for Cancer Research, University of Lausanne, Lausanne, Switzerland
1 shared publication - Janani Durairaj
Biozentrum, University of Basel, Basel, Switzerland
1 shared publication - Jeffrey W. Bode
ETH Zurich
1 shared publication - Jiling Han
ETH Zurich
1 shared publication - Julian van Gerwen
Charles Perkins Centre, School of Life and Environmental Sciences, University of Sydney
1 shared publication
Latest Publications
- Open Access
Capabilities, specificity gaps and training-data dependence of AlphaFold3 across diverse application areas
Structure prediction models have moved from single proteins to assemblies that include diverse biomolecules and their modifications. AlphaFold3 (AF3) and related models extended structural modelling via an all-atom framework, opening many new potential applications in structural biology. We evaluate how well the new capabilities of AF3 translate into application tasks in diverse areas: prediction of ubiquitinated protein structures, T-cell receptor (TCR)-epitope recognition, antibody-antigen complexes, protein-RNA and protein-lipid interactions. We find that, while AF3 can perform well in favourable settings, this performance is uneven across applications. In RNA-target predictions, the model confidence fails to separate genuine from decoy interaction partners and in several tasks accuracy depends on the presence of related complexes in the training set. Taken together, our assessment is more cautious than for AF2, whose gains in modelling monomers and complexes were clear and broadly generalisable. AF3’s extension to new biomolecule types shows less consistent performance and generalisation. AF3 can be a powerful tool for hypothesis generation and prioritisation, but its predictions and use of confidence metrics will depend strongly on the specific application area and must be interpreted with respect to training-set overlap. We expect that the benchmarks provided here will serve for testing of future developments in the structure prediction field.
Open AccessCapabilities, specificity gaps and training-data dependence of AlphaFold3 across diverse application areas
Structure prediction models have moved from single proteins to assemblies that include diverse biomolecules and their modifications. AlphaFold3 (AF3) and related models extended structural modelling via an all-atom framework, opening many new potential applications in structural biology. We evaluate how well the new capabilities of AF3 translate into application tasks in diverse areas: prediction of ubiquitinated protein structures, T-cell receptor (TCR)-epitope recognition, antibody-antigen complexes, protein-RNA and protein-lipid interactions. We find that, while AF3 can perform well in favourable settings, this performance is uneven across applications. In RNA-target predictions, the model confidence fails to separate genuine from decoy interaction partners and in several tasks accuracy depends on the presence of related complexes in the training set. Taken together, our assessment is more cautious than for AF2, whose gains in modelling monomers and complexes were clear and broadly generalisable. AF3’s extension to new biomolecule types shows less consistent performance and generalisation. AF3 can be a powerful tool for hypothesis generation and prioritisation, but its predictions and use of confidence metrics will depend strongly on the specific application area and must be interpreted with respect to training-set overlap. We expect that the benchmarks provided here will serve for testing of future developments in the structure prediction field.
- Open Access
- 11 September 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Comprehensive synthesis of K48/K63 ubiquitin pentamers by graph-empowered automation
Linear, branched, and heterotypic ubiquitin (Ub) chains regulate essential cellular processes, yet their structural and functional analysis remains limited by the difficulty of preparing defined, homogeneous samples by chemical synthesis or biochemical assembly. We established a fully automated platform for the comprehensive synthesis of all 56 K48/K63-linked Ub tetramers and pentamers. This is achieved by iterative cycles of K48 or K63-specific chemoenzymatic conjugations of chemically synthesized Ub monomers bearing a new, fast-cleaving lysine protecting group compatible with folded proteins. A graph-based digital encoding of the Ub chain topologies enabled systematic enumeration and filtering of reaction pathways, guiding the selection of viable synthetic routes. This computational system generates reagent maps and executable code to direct iterative cycles of deprotection and conjugation on a standard, laboratory liquid handling robot. All 14 Ub tetramers and 42 Ub pentamers were obtained with high fidelity, as validated by SDS-PAGE and mass spectrometry. This work establishes a general framework for automated protein conjugate synthesis and provides on-demand access to the complete space of K48/K63 branched ubiquitin chains for biological and structural studies.
- 11 September 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessComprehensive synthesis of K48/K63 ubiquitin pentamers by graph-empowered automation
Linear, branched, and heterotypic ubiquitin (Ub) chains regulate essential cellular processes, yet their structural and functional analysis remains limited by the difficulty of preparing defined, homogeneous samples by chemical synthesis or biochemical assembly. We established a fully automated platform for the comprehensive synthesis of all 56 K48/K63-linked Ub tetramers and pentamers. This is achieved by iterative cycles of K48 or K63-specific chemoenzymatic conjugations of chemically synthesized Ub monomers bearing a new, fast-cleaving lysine protecting group compatible with folded proteins. A graph-based digital encoding of the Ub chain topologies enabled systematic enumeration and filtering of reaction pathways, guiding the selection of viable synthetic routes. This computational system generates reagent maps and executable code to direct iterative cycles of deprotection and conjugation on a standard, laboratory liquid handling robot. All 14 Ub tetramers and 42 Ub pentamers were obtained with high fidelity, as validated by SDS-PAGE and mass spectrometry. This work establishes a general framework for automated protein conjugate synthesis and provides on-demand access to the complete space of K48/K63 branched ubiquitin chains for biological and structural studies.