Biomedical Science

The HANAMI project’s biomedical research will focus on molecular simulations related to respiratory diseases.

Molecular dynamics simulations are used as a complement to experimental methods, offering high space and time resolution for better understanding molecular processes. Researchers will enhance data-driven simulations by combining information from various experimental techniques to guide state sampling, using the European GROMACS code and the Japanese GENESIS code.

To tackle these challenges, a combined approach using CFD, HPC, and machine learning (ML) will be needed, utilizing both European (m-AIA) and Japanese (CUBE) codes. Internally, the focus is on supporting surgeons in treating respiratory diseases. An automated surgery planning tool, currently limited, will be improved with European and Japanese resources and techniques. This can help increase the success rate of surgeries, such as those for nasal septum deviation, which currently stands at just 55%. Externally, the focus will be on improving airflow in waiting rooms, including managing the spread of droplets from coughing. The goal is to optimize the environment, such as ventilation design. From the internal perspective, optimization will be enhanced by integrating jointly developed ML algorithms.
To address these challenges, a combined approach of CFD, HPC, and machine learning (ML) will be required from an internal and external perspective (with the European code m-AIA and the Japanese code CUBE). The internal aspect is dedicated to assisting the surgeon in the decision-making process to treat respiratory diseases. An automated surgery planning tool with limitations at this stage must be improved with European and Japanese resources and techniques. The results can be expected to improve success rates of surgeries (in interventions to correct a deviation of the nasal septum, this figure is only 55%). The external aspect will focus on breathing in waiting rooms, including spreading droplets from, e.g., coughing. The goal is to optimize the interior design, e.g., ventilation inlets. From the internal perspective, optimisation efficiency will be improved by integrating jointly developed ML algorithms.

In the scope of Biomedical Science area, HANAMI will work on three different projects:

1
Exascale Electrostatics & Machine Learning to Enable Molecular Dynamics of Cell-size Systems
2
Development of Genome Analysis Pipelines for Personalized Medicine
3
Personalized Medicine on Macroscopic Level: Respiratory Flows and Infection Risk Analyses

Exascale Electrostatics & Machine Learning to Enable Molecular Dynamics of Cell-size Systems

The molecular life science pillar strategy aims to expand existing collaborations on software and hardware in molecular dynamics simulations, focusing on long-term development of the GROMACS (EU) and GENESIS (RIKEN) codes. It also involves coordinating with the simulation hardware from the RIKEN MD-GRAPE project. These two codebases complement each other: GENESIS specializes in scaling on Fugaku’s high-end ARM CPUs, while GROMACS is optimized for various GPU (NVIDIA, AMD, Intel) and CPU architectures. The collaboration will use both codes as benchmarks and optimization targets in post-Fugaku projects, offering a unique co-design opportunity for both European and RIKEN teams.

Development of Genome Analysis Pipelines for Personalized Medicine

This project builds on the ongoing collaboration between Japan and Europe within the International Cancer Genome Consortium (ICGC), the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Project, and the ICGC-Argo project. It also draws on work developed by Prof. David Torrents (BSC) during his 2022 sabbatical in Japan, where he collaborated with Japanese groups on genome analysis workflows, a key part of this project. Additional contacts made during Prof. Barillot’s visit to RIKEN in collaboration with PRAIRIE will also be explored.
The goal of this project is to create mechanistic models for individual patients based on their genomic makeup and treatment responses. These models will be reviewed by field experts to identify potential treatments and biomarkers. By developing infrastructure for personalized medicine, this project aims to create practical tools that will support clinicians and genome scientists, ultimately improving tailored cancer treatment strategies.

Personalized Medicine on Macroscopic Level: Respiratory Flows and Infection Risk Analyses

This project builds on a long-standing collaboration between the Jülich Supercomputing Centre (JSC), Forschungszentrum Jülich, and the RIKEN Center for Computational Science (R-CCS), under the Joint Laboratory for Extreme Scale Computing (JLESC). In the first project, "Comparison of Meshing and CFD Methods for Accurate Flow Simulations on HPC systems," the m-AIA and CUBE codes from RWTH Aachen University and RIKEN R-CCS were ported and benchmarked on Jülich and RIKEN machines. The recent project, "Deep Neural Networks for CFD Simulations," focuses on developing AI methods for engineering applications, particularly in biomedical computations, such as respiratory flow analysis for personalized medicine, part of CoE RAISE.
The first goal of this project is to develop AI-assisted automated methods for running large-scale computational fluid dynamics (CFD) simulations, including analyzing results for individual patients. The second goal is to use AI methods and novel hardware technologies at JSC and RIKEN R-CCS to speed up computations. The third goal is to provide an AI-assisted surgery planning tool for surgeons as a demonstration. The methodologies developed will also help achieve the final goal: assessing the risk of exhaled infectious viruses. All implementations will enhance the capabilities of the m-AIA and CUBE codes.

Meet the Team

WP5

Alfonso Valencia

Barcelona Supercomputing Centre (BSC)

WP5

Andreas Lintermann

Forschungszentrum Jülich (FZJ)

WP5

Erik Lindahl

Royal Institute of Technology (KTH)

WP5

Florence Tama

WP5

Wolfgang Schröder

WP5

Julio Saez-Rodriguez

Keep exploring
our project!

Results

Learn more