Cell Analysis | Molecular Mapping | IND
Biometrics | Gene Sequencing | Membrane Proteins
A few days ago, the famous structural biology scientist Yan Ning announced that he would resign from his teaching position at Princeton University in the United States and return to China to serve as the founding dean of the Shenzhen Academy of Medical Sciences.
Yan Ning, born in Zhangqiu, Shandong Province, was born in November 1977. In 2007, he was employed by Tsinghua University School of Medicine as a professor and doctoral supervisor. In 2012, he was qualified as a tenured professor. In 2013, he became a Bayer chair professor. In 2017, he was hired by Princeton University and became the first Shirley Tillman Lifetime Chair Professor.
She is mainly committed to the structure and mechanism of transmembrane transport proteins. She has revealed for the first time in the world a series of transmembrane proteins with important physiological and pathological significance, such as human glucose transporter, eukaryotic voltage-gated sodium ion channel and calcium ion channel. The atomic resolution structure provides a molecular basis for understanding the pathogenic mechanism of related diseases and drug development.
Since 2009, Yan Ning has published nearly 80 academic research papers as a corresponding or co-corresponding author, 33 of which were published in "Cell", "Nature" and "Science".
Yan Ning won the Science/AAAS and GE Healthcare "Young Scientist Award" (North America) in 2005; the first International Young Scientist Award from HHMI in the United States and the "China Outstanding Young Female Scientist Award" in 2012; Technology Progress Award; In 2015, he won the Young Scientist Award of the International Protein Society, the Sackler International Biophysics Award, and was hired as the Changjiang Chair Professor (terminated in 2018); in 2016, he became the first Gordon Research Conference from mainland China The Alexander Cruickshank reporter; in 2018, won the Asia and Oceania Federation of Biochemists and Molecular Biologists (FAOBMB) "Excellent Research Award"; in 2019, won the International "Women in Science Award" issued by the Weissman Institute in Israel; in 2021 Received the Anatrace Membrane Protein Research Award from the International Biophysical Association.
What is Yan Ning's sudden return to China?
In 2014, Yan Ning led a team with an average age of less than 30 years old. It took only 6 months to analyze the crystal structure and working mechanism of the human glucose transporter GLUT1 for the first time, successfully overcoming the molecular structure scientists in the world. A difficult problem that has lasted 50 years, this achievement has also made Yan Ning gain countless honors. However, such research results did not arouse widespread concern of Internet public opinion. The two biggest controversies, one was leaving and the other was coming back. The main controversy of Yan Ning's return to China this time is: "It is said that she has been academically bankrupt in the United States and cannot continue to work, because her job has been replaced by AI." Many keyboard warriors who do not understand science and AI at all embrace A feeling of schadenfreude. An article popular on the Internet summed up the whole incident with extremely exaggerated and simple sentences at the beginning.
Let’s first look at another protagonist in this controversy – what is “AlphaFold”? What did you do again?
AlphaFold can be said to be the AlphaGo of the biological world, and its contributions have also been called "epoch-making progress" by scientists in this field. Also from the DeepMind company AlphaFold, like AlphaGo, which swept the top human professional Go players, defeated the rest of the participants in the International Protein Structure Prediction Competition (CASP) in 2018 (first participation) and 2020 respectively. , and caused a sensation in the field of biology.
After DeepMind released the source code of AlphaFold, the biological community also set off a wave of "AlphaFold frenzy" research.
In the 50 years before the birth of AlphaFold, scientists required a lot of trial and error, consuming a lot of time and energy, using experimental methods such as cryo-electron microscopy, nuclear magnetic resonance or X-ray crystallography to determine the shape of proteins in the laboratory.
In 1972, Nobel Prize winner Christian Anfinsen proposed that in theory, the amino acid sequence of a protein should be able to completely determine its 3D structure. This hypothesis led scientists to explore computational methods to predict the 3D structure of proteins based on amino acid sequences.
However, such explorations face extremely significant challenges, and early attempts in the 1980s and 1990s to predict protein structures using computers were unsuccessful. The emergence of AlphaFold has changed this situation. In 2021, DeepMind announced that it has used AlphaFold to predict the structure of almost all proteins in the human body, as well as the complete "proteome" of 20 other heavily studied organisms, including mice and E. coli, with a total of 365,000 structures.
And this year, DeepMind plans to release a total of more than 100 million structural predictions—equivalent to nearly half of all known proteins, and hundreds of experimentally resolved proteins in the Protein Data Bank (Protein Data Bank) structure database. times as much. Yan Ning's achievements have overcome a problem that has plagued the biology community for 50 years, and AlphaFold's achievements have also solved a major challenge that has plagued the biology community for 50 years.
So some people say: You see, Yan Ning and her team have been researching for so long to come up with a structure. AlphaFold can predict hundreds of millions at once, which is equivalent to subverting the entire game. Then people like Yan Ning and the others Of course, he was laid off and had nowhere to go.
Here are a few things to think about:
1. The irrational numbers and imaginary numbers that elementary school students know now took people thousands of years of thinking to appear in history. We cannot say that those mathematicians made things that elementary school students understand.
2. The capabilities of AI come from the data accumulated by predecessors. The reason why AlphaFold can do this is precisely because scientists have done countless theoretical and experimental work before, and it has a very good foundation. AI can only reach such heights by standing on the shoulders of many giants. Without the previous results of these scientists, it is impossible for AI and machine learning to learn to do biological research by themselves.
3. Scientists do not only do this one thing, they will be replaced if AI can do it. Because the AlphaFold research is so complex, if AI can solve it, it will precisely liberate these scientists from a large number of repetitive and tedious experiments, allowing them to study more and more interesting topics. In fact, it is equivalent to the fact that we invented bicycles and cars, and then they can help humans walk. This does not mean that humans do not need to walk, but that people save this time and provide humans with greater freedom. Let people do more meaningful things.
Fourth, Yan Ning didn't want to return to China because she couldn't survive abroad. A top scientist like Yan Ning has published so many "Nature", "Science", and "CELL", it is impossible to just do repetitive experiments with a dull head. People who have never published a paper think out of thin air that they can publish a paper for the top journal just by staying up late and doing repetitive work.
Why Shenzhen Academy of Medical Sciences can attract Yan Ning to serve
According to a report released by Shenzhen, the Shenzhen Academy of Medical Sciences was established by the municipal government and registered as a public institution organized by the municipal government. It implements a system of dean responsibility under the leadership of the party committee; it has no fixed staffing and levels, and implements a socialized employment system. The board of directors is the decision-making body of Shenzhen Academy of Medical Sciences; the dean is the legal representative of Shenzhen Academy of Medical Sciences, recruiting from all over the world, appointed by the board of directors, and implementing a tenure system.
According to reports, Shenzhen Academy of Medical Sciences will mainly build "four platforms and one think tank" in accordance with the requirements of the new mechanism, and strive to become a world-renowned medical research institution by the middle of this century.
That is to say, Shenzhen Academy of Medical Sciences is a brand-new medical academy supported by the state. The so-called "new mechanism" includes two aspects.
1. New positioning
That is to say, Shenzhen Academy of Medical Sciences is not only a pure research institution, but according to the official statement, it is a "scientific research organization that organizes scientific research". Its core functions are to undertake public management and service functions in medical science and technology research. In addition, it is also necessary to lead the development of medical science and technology in Shenzhen. To this end, the Shenzhen Municipal Government has also established the "Shenzhen Medical Research Special Fund" and entrusted the Shenzhen Academy of Medical Sciences to conduct professional management.
2. New mechanism
Indeterminate staffing, indeterminate level, set up posts independently, follow the principles of council governance and academic autonomy. For scientific research talents including deans, a market-based salary and social employment system is implemented. In September last year, the Shenzhen Health and Health Commission issued a wave of recruitment for management positions at the Shenzhen Academy of Medical Sciences.
3. Government policy support
The new mechanism is not only reflected in personnel. Although the Shenzhen Academy of Medical Sciences is registered as a public institution directly under the Shenzhen Municipal Government, it is essentially a statutory institution and implements "one hospital, one law". Specifically, the government will issue the "Administrative Measures for Shenzhen Academy of Medical Sciences", which will allow it to run its own institute according to law without taking into account the system of traditional public institutions.
4. Shenzhen Academy of Medical Sciences has strong financial resources
The first is special government funding, which is the "Shenzhen Special Fund for Medical Research" mentioned above. Documents released in May showed the government allocated $28.48 million to its budget in 2022. At the same time, the Shenzhen Academy of Medical Sciences will also set up a joint fund, accept charitable donations, introduce venture capital, and gradually explore the establishment of the "Guidelines Fund for Health Science and Technology Innovation in the Guangdong-Hong Kong-Macao Greater Bay Area". In addition, another major source of funding for the Academy of Medical Sciences is the transformation and production of drugs and equipment, and the profits from the transformation are directly fed back to itself.
5. Future development plan of Shenzhen Academy of Medical Sciences
Regarding the future development plan of Shenzhen Academy of Medical Sciences, there are two points to sum up.
1) aggregate resources
Shenzhen Academy of Medical Sciences is equivalent to a collaborative innovation platform for medical science and technology, which solves the problem of scattered allocation of domestic medical science and technology resources, avoids cross-wasting of resources, and inefficient use of scientific research funds.
2) Help transform scientific research results
Shenzhen will allow scientific researchers to hold shares in transformation projects through "technical investment", directly participate in the transformation process of scientific and technological achievements, and increase enthusiasm for transformation.
In addition, Shenzhen Academy of Medical Sciences will also transform enterprises through "shareholding" through angel investment and other forms, and gradually transition from a single scientific and technological research and development to a scientific research and industry hybrid. According to the construction plan of Shenzhen Academy of Medical Sciences, Shenzhen Academy of Medical Sciences will be basically completed in 2025.
Yan Ning also said:
"One of the important missions of Shenzhen Academy of Medical Sciences is to closely integrate research, medicine, and medicine, and open up a smooth end-to-end connection from the hospital bed to the laboratory, to the pharmaceutical company, and finally back to the hospital bed. I hope that Shenzhen Academy of Medical Sciences can not only produce A number of original scientific research breakthroughs can also explore a scientific and reasonable mechanism, which can effectively help everyone realize the transformation of scientific research results while ensuring that researchers have specialization in their skills and focus on scientific research. ”
Blue Ocean Brain brings liquid-cooled servers to help the development of life medicine
The single-cell genome research technology center of a college (referred to as "the center") aims to establish standardized and automated engineering technology, improve the level of single-cell structure analysis, determine the three-dimensional structure from protein molecules to whole cells with high precision, and on this basis Reveal the functions of proteins and their complexes, prepare proteins/antibodies on a large scale, and build a core base for protein science research with world-class level and comprehensive demonstration functions.
As far as life science research projects are concerned, the amount of data involved in each project is as small as hundreds of terabytes. For projects with a long time period and a wide range of fields, the future data demand may be at the PB level. In addition, the center needs to consider supporting a variety of life science research projects, and among them, different applications have different requirements for high-performance platform computing environments, such as gene sequencing requires high I/O performance and large memory consumption, while molecular dynamics research in addition to In addition to I/O performance, high network and concurrent processing capabilities are also required.
1. Current challenges in the field of life sciences
1) The amount of data has increased by more than 10 times, and the computing power must "keep up"
The cryo-electron microscopy technology adopted by the research team has made revolutionary progress in the past two years. Specifically, the camera technology has achieved a leap forward, and the ability to collect data has increased by more than 10 times, or even hundreds of times, so that the source data for protein structure research The growth is geometric progression, which requires the center to comprehensively improve data processing and computing capabilities in the later stage.
2) Urgent need to simplify management to ensure service quality
With more and more life science research projects, how to allocate resources according to the individual needs of different projects and researchers, recycle resources in a timely manner, realize centralized and unified management across the entire high-performance resource pool, simplify maintenance management, and reduce the burden on operation and maintenance personnel, It is a common problem faced by high-performance computing platforms for scientific research.
3) TCO remains high
Life science research has quickly become a national strategic development direction, leading to a rapid increase in research projects and interdisciplinary research needs. The low utilization rate of traditional tiered computing and storage resources leads to a rapid increase in new costs. In addition, energy consumption has become an insurmountable "high wall" that hinders the expansion of high-performance computing centers.
4) Network performance should not be a hindrance
As the key to ensure the normal operation of high-performance clusters, high-performance networks undertake important connection tasks. With the continuous improvement of single-node computing and storage performance, high-performance users need 10G, 40G, 100G, and InfiniBand network options to meet different high-performance computing needs.
2. Features of the solution
Based on the fusion architecture, Blue Ocean Brain helped a college's single-cell genome research technology center build a distributed high-performance platform with 250 physical computing nodes, 5,000 computing cores, a total storage capacity of 1.92PB, and a theoretical computing capacity of 208Tflops. Centralized and unified management across 20 converged architectures is realized through Luster technology.
1) 4.1TFLOPS/U calculation density, 4 times performance improvement
Configurations can be tailored for different projects. Among them, the high-density computing node supports the 14-core Intel® Xeon™ E5-2600v3 processor, has a density of 224 computing cores in 2U, and the computing performance density of a single U space reaches an industry-leading 4.1TFLOPS, and supports 64 DIMMs at the same time High-density memory ensures high performance and low latency performance requirements. In addition, InfiniBand interfaces are supported, ideal for workloads requiring ultra-low latency. With the guarantee of strong computing power, the computing efficiency is increased by 3-4 times, and the computing tasks that were completed in the past 4-5 days can be completed in one day.
2) Simplify the monitoring and management of high-performance resource pools
Different system configurations can be customized according to project requirements, and 20 FX systems can be centrally monitored and managed through the Chassis Management Controller (CMC). In addition, agentless lifecycle management and one-to-many remote management functions can ensure that BIOS and firmware program updates will not affect business stability, and improve the efficiency of lifecycle management of computing nodes in the system. Moreover, when expanding the server, IT personnel can issue configuration files to enable the system to automatically update the BIOS and firmware programs, avoiding the tedious process of repeatedly inputting configuration parameters, reducing system failures caused by manual input errors, and simplifying management, operation and maintenance. Reduced management costs.
3) TCO is reduced by about 20%
Integrated deployment with automation, high density and low energy consumption, and centralized and unified management can reduce the TCO of the center by about 20%. Among them, Blue Ocean Brain will connect the server, storage and 1G0b network through the motherboard, and form a fusion all-in-one machine through modular design. At the same time, it will provide shared slots for heat dissipation, power supply, network, management and PCIe expansion, reducing the footprint and energy consumption of the data center. , help center get good value for money.
4) High-speed network guarantees platform I/O performance
Blue Ocean Brain provides the center with a 40G high-performance network. On the basis of maintaining cost advantages, it provides users with stable network performance and guarantees high performance and low latency requirements.
5) Break the original server heat dissipation method and use liquid cooling to dissipate heat
The Blue Ocean Brain liquid-cooled server system breaks through the traditional air-cooled heat dissipation mode and adopts a mixed heat dissipation mode of air-cooled and liquid-cooled - the main heat source in the server, the CPU, is cooled by a liquid-cooled cold plate, and the remaining heat sources are still cooled by air-cooled methods. Through this hybrid cooling method, the heat dissipation efficiency of the server can be greatly improved, and at the same time, the power consumption of the main heat source CPU heat dissipation can be reduced, and the reliability of the server can be enhanced. After testing, the annual average PUE value of data centers using liquid-cooled server supporting infrastructure solutions can be reduced to below 1.2.
3. Customer benefits
1) The Blue Ocean Brain HPC high-performance computing and AI platform has become a high-performance, multi-functional, professional cutting-edge computing platform, especially in the aspect of AI deep learning, providing efficient computing support for biological research inside and outside the school. At the same time, it provides computing services for multiple research groups such as computational biology, deep learning, and gene sequencing. Including off-line processing of sequencers, sequence search and comparison analysis, molecular dynamics simulation, computer-aided drug design and molecular docking, and calculation of biological networks.
2) Fully support the development of deep learning-based molecular graph coding and deep learning-based traditional Chinese medicine prescription system. R&D personnel can use HPC high-performance computing and AI platforms to develop deep learning codes based on three-dimensional molecular maps, and carry out TCM diagnostic prescriptions based on deep learning. The multi-task molecular prediction model is composed of convolutional neural network or recurrent neural network. Cross-validation is used to tune and validate parameters, and external data is used to test and evaluate the model. At the same time, key information is mined from the predictive model. At the same time, a large amount of prescription compatibility information is learned through convolutional neural network or recurrent neural network, and then the main drug is used to generate the adjuvant obtained by semantic automatic correlation analysis, thereby generating a new prescription. Blue Ocean Brain's HPC high-performance computing and AI platform provides efficient parallel computing resources, which greatly accelerates the training speed of the model, so that the final task can be completed within an effective time.
3) Support the ab initio drug design based on chemical fragments, which has an important role in promoting the treatment of diseases and the understanding of biological functions. The traditional drug screening process is time-consuming and costly, resulting in inefficiency in the entire drug design and discovery process. In order to speed up the process of drug design and discovery, researchers used this platform to gradually develop the method of molecular de novo design, and achieved good results. Through the combination of Monte Carlo tree search and neural network model, the researchers realized the search of huge chemical space and the sampling of the optimal structure, quickly completed the complete ab initio drug design process, and explored the protein pocket characterization and scoring functions.
4) Use the deep learning framework to build a deep learning model, strengthen the training of the learning model, realize the training and testing of the deep learning scoring function model, and train the model. For the molecules generated by the model, the synthesis, toxicity, and physicochemical properties of the molecules were analyzed by clustering to select the appropriate molecules.