Biomedical Data Analytics and Supercomputing Core

Guangzhou Branch of the Supercomputing Center of CAS

1、 Overview

The Guangzhou Sub-centre for Supercomputing of the Chinese Academy of Sciences (hereinafter referred to as the ‘Supercomputing Centre’) relies on the Biomedical Data and Supercomputing Centre of the Guangzhou Institutes of Biomedicine and Health of the Chinese Academy of Sciences (hereinafter referred to as the ‘GIBH’) for its construction and operation.

The total area of the supercomputing centre is 117m2, including office area of 57m2, machine room building area of 60m2, 12 water-cooled column cabinets, and a total of 117 nodes have been deployed for data computing. Existing parallel storage resources 7PB, aggregation capacity of more than 200TFlops, with 100M high-speed technology network bandwidth and 1000M China Mobile dedicated bandwidth, for the management of data resources and open and shared services to provide a highly reliable and stable infrastructure operating environment.

More than 1,000 application software for cellular genealogy, genealogy technology and equipment, development and regenerative medicine, digital biomedicine, pharmaceutical innovation and other multidisciplinary fields are centrally deployed to carry out large-scale scientific computation, data simulation, experimental validation and related data visualisation services.

2、 Services

·High-performance computing services

We provide hardware and software services for supercomputing (high-performance computing, parallel computing) to users inside and outside the Guangzhou Healthcare Institute for scientific research and teaching. Diversified hardware configurations of supercomputer systems (high-performance CPU and storage system, low-latency high-speed computing communication network, etc.), equipped with many open-source and commercial computing software, and full-time technical staff to provide users with a full range of support and services through multiple channels, such as telephone, e-mail and instant messaging, to provide users with a good supercomputing environment.

·One-stop data analysis services

We have gathered thousands of bio-multi-omics and innovative drug analysis toolkits, and more than 50 commonly used visualisation application script toolkits in the field, to build a cellular mapping data mining platform, which can carry out research on multi-omics of genomics, transcriptomics, epigenomics, single-cell sequencing and so on.

Digital Twin 3D Visualisation

The Supercomputing Centre is committed to the independent development of advanced visualisation tools, through the use of WebGL, WebGPU, Threejs and other computer graphics libraries, to achieve the rapid construction of 3D dynamic scenes on the Web, 3D interaction, animation rendering, as well as efficient management and scheduling of resources and other functions. The Supercomputing Centre is equipped with the technical capability of 3D modelling, and is able to use Blender and other design software to draw 3D digital models of site parks, buildings, equipment and instruments. At the same time, it is able to use generative artificial intelligence technology to build complex and diverse organ, tissue, cellular and subcellular structure models. The Supercomputing Centre uses the above tools and technical capabilities to serve the development of digital human interaction applications and the construction of the Digital Twin Intelligence Institute.

·LLM Applications and Artificial Intelligence

Supercomputing Centre integrates and deploys a variety of multimodal big models, such as Llama 3.1, ChatGPT, GPT4, Qwen2, etc., which can be selected by users according to their preferences, providing users with one-stop AI services.The Supercomputing Centre has carried out application development based on the big models, hoping to assist the various processes of scientific research and management through the big models, and serve the construction of the digital management system with the core of the business system.

Resources

4、Contact

Center for Scientific Data of GIBH,CAS

1、 Overview

The Scientific Data Centre (hereinafter referred to as the ‘Data Centre’) of the Guangzhou Institute of Biomedicine and Health, Chinese Academy of Sciences (hereinafter referred to as the ‘GIBSH’) was constructed and operated by the GIBSH as a supporting unit, and was accredited as one of the first scientific data centres by the Chinese Academy of Sciences in 2021. In 2021, it was accredited as one of the first scientific data centres by the Chinese Academy of Sciences.

Aiming at the actual needs of scientific data storage, management and use, as well as the major problems of ‘data silo’ and ‘data sovereignty’, the data centre focuses on the following areas of cytomics research: genome, transcriptome, transcriptome, transcription, and transcription. Focusing on the genome, transcriptome, epigenome, proteome, metabolome, immunome and other data resources involved in cellular genomics research, we will focus on the construction and service of a one-stop platform for data management and data application, and carry out four aspects of work, including data collection, data remittance, data analysis, and data visualisation, etc., so as to provide the services of remittance management and sharing of scientific data requested by the Institute and domestic and foreign scientific research projects and theses and journals and to promote open sharing of scientific data and guarantee the safety and reliability of scientific data. It promotes the open sharing of scientific data, ensures the safety and control of scientific data, and supports national scientific and technological innovation and economic and social development.

2、 Services

·Data Archive

The total number of archive thesis-related data exceeds 100, and the number of national science and technology program projects exceeds 40. Accumulated biological multi-omics data exceeds 1.5PB.

3、Resources

·Data resources

Stem cell and biomedical science and technology cloud platform and stem cell and metabolic disease database have been set up, and 35 data repositories, total data volume of 170TB, and number of data entries of 3.7 million have been released.

4、Contact