Ingemar André

The paper develops a transformer model to predict codon sequences from amino acids, improving rare codon prediction and linking sequence variation to protein function.

The paper designs ancient proteins using a limited amino acid set and shows they are stable and mutation-resilient.

This study introduces a mutation-selection model to predict site- and amino-acid substitution rates from amino acid sequences, showing it outperforms standard phylogenetic methods and correlates with empirical Bayes estimates.

The paper investigates how single synonymous substitutions affect protein folding efficiency in E. coli, showing that rare codons and protein structure significantly influence this process.

The paper analyzes protein-protein interfaces in T=1 virus capsids to determine their structural and energetic properties compared to non-viral homomers and their role in self-assembly.

The paper shows that averaging site-specific mutation rates across homologous protein structures improves predictions of site-rates from structure.

This study engineers multivalent Fc-displaying virus-like particles to block FcγR activation, showing that higher valency reduces receptor activation and enhances inhibition of immune complex-mediated phagocytosis.

This paper combines AlphaFold with symmetric docking to accurately predict structures of symmetrical protein assemblies from sequence.

This study uses an all-atom simulator to show how fitness functions affect mutation rates and residue covariation in protein evolution.

This paper introduces EvoDOCK, a protein-protein docking algorithm that combines global search and local optimization to improve accuracy and speed using all-atom energy functions.

This paper introduces the JaLT algorithm to sample conformational transitions in protein loops, demonstrating its ability to generate realistic pathways and identify energy barriers in the MET20 loop of E. coli dihydrofolate reductase.