In iTHEMS biology seminar on August 5th, Dr. Ai Niitsu (RIKEN Theoretical Molecular Science Lab.) gave a talk about the membrane peptide design and computational modeling.

First, she mentioned the recent machine learning approach for the prediction of the protein structures. Then, she pointed out that the prediction of the membrane related structure and stoichiometry of protein complexes is still challenging.

Next, she explained the computational design of the coiled-coil peptides in membrane. The amino acid sequence of the peptide is determined so that the interchain interaction is satisfied. As a result, she obtained some stable transmembrane assemblies of the peptide.

Along with the modelling of the peptide assemblies and their conductance estimate, she experimentally measured the conductance of the peptide channel in membrane by imposing various voltages. The high voltage induces the multi-conductance state, which differs from the low voltage case where a stable single channel was observed. Based on further computational modelling and experiments, that multi-conductance state was suggested to come from the change in the peptide structures and stoichiometry of a single pore.

Her computational method and experimental results attract the audience interests and induce various discussions. Thank you very much for great talk, Ai!

Reported by Hiroshi Yokota

関連イベント