research

      home             profile             research             publications             Japanese ver.      

Research interests

My research is aimed to achieve a new vision of life science research by harnessing robotics, large-scale measurement, and data science.

List of Current Research Topics

1. Development of microbial breeding strategies through omics analysis, high-throughput experiments, and data-driven science

The use of microorganisms to produce materials such as biofuels and chemical raw materials is increasing in attention in response to the demand for a sustainable society. In this process, breeding of resistant microorganisms against various stress such as resulting from the accumulation of products is necessary. I have been working on the breeding and control of microorganisms by utilizing the ability of evolution of living systems.

In my doctoral program, I began research using a methodology called adaptive laboratory evolution (ALE). ALE is a method of inducing evolution in the laboratory by utilizing the evolutionary ability of organisms to mutate and select against their environment (see the right figure). Omics measurement technologies such as next-generation sequencers make it possible to analyze mutations and changes in intracellular conditions that occur during the evolutionary process. Even if the physiological mechanism necessary for adaptation to the environment is unknown, organisms themselves can make such adaptation happen, and omics measurement can elucidate the cause of such adaptation (e.g. ref. [1-1], [1-2], [1-3]).

On the other hand, ALE requires a long experimental period and a high workload for the experimenters, making the process a bottleneck in the execution of research. Therefore, I have developed an automated system that can perform fully automated and high-throughput microbial passaging for ALE in RIKEN (ref. [1-4], movie). This not only solves the human resource problem, but also enables the development of data-driven research that finds useful findings and laws from experimental data under various experimental conditions.

By using high-throughput automated system, ALE under multiple conditions, and omics analysis, we have been able to (i) breeding stress-tolerant E. coli (e.g. ref. [1-5], [1-6], [1-7]), (ii) analyze the emergence mechanism of drug-resistant bacteria (e.g. ref. [1-8], [1-9]), (iii) improvement of the productivity of growth-coupled linked strain (e.g. ref. [1-10], [1-11]). These findings can be utilized for rational breeding of useful microorganisms and development of methods to inhibit the emergence of drug-resistant bacteria.

Recent developments in AI technology and the expansion of computational resources are remarkable, and have already demonstrated great potential in areas such as imaging analysis or prediction of macromolecular structures. On the other hand, obtaining sufficient data to exploit the advantages of AI is still challenging for complex problems, such as microbial culture processes and stress responses, for example, which involve complex elements such as not only the behavior of the cells themselves, which consist of thousands or more genes, but also their interactions with the external environment. We are planning to engage in AI-driven research by combining the data-driven approach we have been performing so far.

2. Advancement of laboratory automation via robotics

The COVID19 epidemics had significant impact on the lifestyles, economic activities, and other aspects of life in the world. Academic research at universities and other institutions is no exception, requiring the establishment of a system for the new era of COVID19. Unexpectedly, the COVID19 highlighted the importance of remote control and automation in the field of research and development. However, various challenges still exist in establishing such systems. For instance, the operation of automated systems still requires numerous human interventions, such as replenishing consumables, cleaning up after use, and troubleshooting, and there are still not many bio-experimental instruments that can be connected to automated platforms. Therefore, I am working on the implementation of a unique laboratory automation system in AIST, which utilizes arm-type robots and associated technologies which have supported the manufacturing industry. 

LabDroid Mahoro, developed by Robotic Biology Institute, Inc., can use the same experimental equipment (centrifuge, vortex mixer, etc.) as used by humans, and can perform complex movements such as tilting and rotating with its dual-arm. I am participating in research projects using Mahoro, taking advantage of my previous experience in laboratory automation. Currently, I am working on the automation of animal cell culture and omics experiments (e.g.,  ref. [2-1]), and the advancement of automated experiments by integrating with high-throughput pipetting machines (right figure) (ongoing research projects [2-1], [2-2]).

3. Bottom-up automation for low-cost implementation and popularization

Laboratory automation is a key technology for data-driven and AI-driven science. However, many of these technologies require expensive or custom-built equipment making them far from widely available to many researchers. Recently, low-cost semi-automated pipetting machines, inexpensive robots developed for STEM education, and IoT devices have emerged. To address this situation, I am trying to implement inexpensive experiment automation systems using these devices and make them available to many researchers by releasing as an open-source and open-platform (see the figure).

Although several examples of home-built systems using educational robots and IoT devices are available on the web, the number of such systems is still small, and it is difficult for biotechnologists to find a way to use them. These examples of home-built systems are just "parts in synthetic biology," in other words, and the expansion of the lineup of such systems is interconnected with the expansion of the population in the field. I believe that a breakthrough to address the current situation can be achieved by the lead in promoting open-source research.

Therefore, I am currently working on bottom-up automation as a PoC on the subject of evolution engineering of proteins. By combining a semi-automated pipetting machine without microplate transfer function and an educational robot (e.g., right movie), I am trying to automate experiments at a lower cost than conventional automation equipment for life science (ref. [3-1], ongoing research project [3-1]).

Web_DIYRobotics.mp4

4. Application and dissemination of bioinformatics technology by wet-lab researcher

Recent advances have led to the emergence of highly comprehensive and analytical approaches, bioinformatics analysis, which is used to extract useful information from huge amounts of biological information, is becoming increasingly important. However, bioinformatics analysis requires advanced expertise, and since there are a wide variety of methodologies depending on the target and purpose of the analysis, it is difficult for beginners to perform. In addition, the lack of trainers and systematic training systems is sometimes experienced in their laboratory environment, resulting in insufficient education for users to conduct their research and forcing them to overcome issues of bioinformatics by themselves. I have also recognized the importance and hurdles through the analysis of omics data I have acquired myself.

Therefore, for the purpose of developing support environments, creating chances for interaction, encouraging wet researchers who could understand and perform bioinformatics, and supporting research activities, the Bioinformatics Consultation Forum (link) has been established in FY2017 through the research division establishment system in The Society for Biotechnology, Japan (SBJ) (link). 

The forum is managed by a group of volunteer wet-lab researchers who also have expertise in bioinformatics. We hold symposia , hands-on seminars, and an online consultation facility (see the figure on the right). The activities of the forum have led to the development of collaborative research, publication of co-authored papers, and editing and writing of books (link, ref. [4-1], [4-2]).

The interaction of researchers inside and outside of the society is important for the development of the field. If you are interested in the activities of this forum, please come to the meeting regardless of whether or not you are a membership of SBJ.

The Bioinformatics Consultation Forum, Japan (link) (Sorry for only Japanese language)

5. Activities to develop and promote for laboratory automation

The implementation and operation of laboratory automation require knowledge and experience in multiple fields, such as biology, mechanical engineering, software engineering, and other interdisciplinary fields (see figure on the right). In addition, since the population of the field itself is small, there are few opportunities to share know-how for implementation and operation, and such field-level information is rarely discussed in academic presentations in the conferences. I have experienced such a situation myself (ref. [5-1]). Therefore, the Laboratory Automation Supplier's Association (link) has been organized for exchanging information among users and developers who are working on, want to join, or are interested in laboratory automation. We hold a monthly workshop and an annual conference. If you are interested in the activities of this forum, please come to the meeting.

Laboratory Automation Supplier's Association (link) (Sorry for only Japanese language)