Acceptance meeting site
From April 17 to 18, the Planning and Finance Bureau of the Chinese Academy of Sciences, together with the Information Technology Work Leading Group Office and the High-Tech Research and Development Bureau organized experts in Beijing on the major project of the Chinese Academy of Sciences Knowledge Innovation Project-"High Performance Computing Research for Protein Science "And 9 topics under the project were accepted on site.
Through four years of hard work, the project has achieved a series of research results in new technologies of high-performance computer systems, new methods of application acceleration in the field, new data of proteome and three-dimensional reconstruction, new algorithms, new software, new collaborative research tools, etc .:
Developed a super-high-performance computer system with super-long No. 1 heterogeneous structure with an aggregate computing power of one hundred billion times, and the acceleration ratio of the single particle three-dimensional reconstruction software EMAN and the electronic tomography three-dimensional reconstruction software ATOM of the cryo-electron microscope reached 10 times. , Effectively supporting biological applications;
Constructed the largest internationally-synthesized protein modification standard peptide library published to date, including more than 260 non-phosphorylated standard peptides and various modified peptides; constructed a protein cryo-electron microscope three-dimensional reconstruction benchmark test data set, total data The amount of 10TB; laid the foundation for the research and evaluation of software and algorithms.
Developed several new algorithms for protein post-translational modification identification and quantitative analysis, developed protein identification search engine pFind, post-translational modification unrestricted search software InsPecT parallel version, quantitative analysis software QuantWiz and other application software. Among them, pFind can find new modifications, and its parallel efficiency on a thousand-core scale system exceeds 80%, which has been widely used.
The new algorithm of three-dimensional reconstruction, such as the adaptive joint algebraic iterative reconstruction algorithm ASART, is proposed. The molecular recognition software Picker, the parallel three-dimensional reconstruction software ParaEMAN, the spherical harmonic function-based three-dimensional reconstruction software ICTIAF, the first domestic electronic tomography 3D Reconstruction software ATOM, frozen electron microscope structural model and density map visualization and analysis software VAT4M. These algorithms and software effectively improve the recognition rate, resolution, reconstruction accuracy and calculation efficiency.
Duckling, a network collaborative scientific research environment software suite, has been developed and has been used as a demonstration application in life science research.
The project achievements have been applied in the research of computational protein science, published more than 60 papers including Structure cover articles, applied for 19 invention patents, and registered 17 software copyrights.
In the study of the HPP architecture, the project team solved the compatibility problem of domestic CPUs with mainstream commercial applications by combining Loongson CPUs and X86 software stacks, and carried out innovative explorations to promote the progress of independent high-performance computing technology. By adopting GPGPU and reconfigurable computing to accelerate the running of bioinformatics, it achieves the purpose of economy and efficiency for specific applications, and explores the breakthrough of traditional HPC growth methods. Biologists have made a number of new discoveries using the new data, new algorithms and new software obtained in the project, and published many high-level articles in many top magazines, which has promoted bioinformatics in algorithms, software, Rapid progress in applications.
An important feature of the project is to cover the entire process of solving physical problems in computational science: including mathematical models, algorithm models and programming, computing platforms, network collaborative scientific research environment, field applications, etc. The interdisciplinary talent team that is both proficient in computing methods and deeply understands the bioinformatics theory has laid a good foundation for the innovative development of the interdisciplinary field of life science and information science in the Chinese Academy of Sciences.
The acceptance team listened to the summary report of each subject and project, reviewed the relevant documents and materials, and after careful inquiry and discussion, they unanimously agreed that each subject and the overall project passed the acceptance.
Dining Room Side Chairs,Leather Side Chairs,Upholstered Side Chair,Mid Century Side Chair
Foshan City Misirui Furniture Co., Limited , https://www.nbmisiruifurniture.com