Laboratory of Informatics of Biological Functions(IMCB) does not recruit students for the academic year 2018

Associate Professor Akio KITAO
Institute of Molecular and Cellular Biosciences(HONGO)
E-mail: kitao{at}
Lab HP


【Key Words】Bio molecule, Molecular dynamics, functional mechanism, cascade type simulation, structure estimation

 Biological activity is maintained by complex interactions between nanoscale functional units. These functional units are biopolymers or supramolecular assemblies, refined over the course of their evolution into biological nanomachines. Our laboratory uses molecular simulations and other methods to investigate the processes required to construct these bio-nanomachines to the point where they carry out functions. Our future goals are the control of bionanomachines and the ability to design new biomachines based on the knowledge we accumulate.

Simulating bio-nanomachine operational Principles

 Exactly how bio-nanomachines form three-dimensional assemblies and fulfill their function is a significant question not limited to basic science, and one that must be answered if we are to apply these machines to the treatment of disease or to custom design our own. Answering these questions will require understanding exactly how they work at an atomic level. We are using supercomputers, PC clusters, and GPGPU machines to run large-scale simulations that allow otherwise impossible in silico observations of atomiclevel processes. This research is uncovering the operational principles behind these machines.

Fig . 1: Bio-nano machines can reachs cales of several million atoms. Bacterial flagellar systems (upper left), proton transport via flag ellarmotors tators (upper right), and membrane penetration processes inviral proteins (belo w).

Bio-nanomachine conformation prediction and modeling

 Predicting the conformation of bio-nanomachines is a prerequisite to large-scale simulations aimed at discovering their function. Our lab is developing ways of applying simulations and bioinformatics to develop models for predicting bio-nanomachine conformation. Representative examples include prediction of protein-low molecular weight complexes, or protein protein complexes. Structure determination has in the past predominantly relied on crystal analysis and solution NMR,but in addition to these we also developing methods for extracting data from neutron scattering and THz spectroscopy experiments to model conformation and dynamics.

Fig . 2: P ro tein protein candidate conformations according to binding free energy calculations from the prediction of complexes.

Developing bio-nanomachine simulations

 We are developing methods of maximizing computer resources, including those of the K computer, allowing efficient molecular simulations of extremely large systems of bio-nanomachines. Specific examples include the development of precise and efficient parallel processing algorithms for large-scale simulations of bio-nanomachines by grid computing,PC clusters,and ultra-parallel computers such as the K computer, and software development for implementation of those algorithms. We are also developing more efficient simulation methods through multi-scale models.
 At our lab, we work toward integrated understanding from a variety of approaches that clarifies complex phenomena like those described above. The graduate students and researchers who have worked here in the past have come from diverse backgrounds including biology, chemistry, physics, pharmacology, information science, and computer science. As we move forward we will increasingly be applying new, multidisciplinary ideas, and therefore we welcome students who are open to breaking new ground that is not limited to the traditional topics of their field.

Collaborative research

 We have developed a number of joint research projects involving close collaborations with both internal and external theoretical and experimental researchers.


The University of Tokyo
Graduate School of Frontier Sciences, The University of Tokyo

Page Top