Recently I saw the WHO’s research roadmap on the new coronavirus epidemic. The epidemic has now become a global outbreak. Strengthening the research and development of drugs and vaccines is also a top priority. Here are a collection of several relevant articles to share with interested researchers and potential collaborators.
The basic principle of machine learning is to find the optimal solution through continuous iteration, similar to Newton's iteration method, through adjustment and iteration in the feasible solution space. By combining a random search similar to a Monte Carlo algorithm, can we find some potential drugs or treatments in the real action space of compounds and drugs against the new coronavirus? The research on new antibiotics in Cell seems to be based on this idea. However, this aspect requires a knowledge base of biology, chemistry, and physics, and it also requires the cooperation of many parties.
IBM is also using supercomputing technology to help fight the new coronavirus. Are some of our domestic computing resources also helping medical research and development institutions? It is hoped that domestic research institutes, universities and other institutions can cooperate to jointly accelerate the development of drugs and vaccines.
It is a pity that the father of YOLO announced his withdrawal from the field of computer vision. How technology can do good is an eternal theme. For example, adding artificial intelligence and simulation technology can quickly identify the working mechanism of the virus and create corresponding drugs and vaccines. This is technology for good. But using it to develop biological weapons is evil.
How to use technology to protect human beings and at the same time protect the environment in which we live so that any subsequent creatures can have room to survive.
Future data technology development can also use agent based modeling. For example, in social governance, such as the prevention and control of infectious diseases, enter some rules, and then let the system evolve by setting some initial values. What key nodes will go wrong in the evolution process? After adding some disturbances, will the system be still stable? And the model can be enriched and evolved by combining actual social management and social practices.
The following is a quote from IBM Summit: The world’s most powerful supercomputer is being used to fight the new coronavirus:
Using Summit, researchers were able to simulate 8,000 compounds in just a few days. By binding these compounds to the viral spike and modeling factors that may affect the infection process, 77 small molecule compounds, such as drugs and natural compounds, were identified. These small molecule compounds have been shown to have the potential to impair the ability of COVID-19 to dock with and infect host cells.
The following is quoted from the CEO of the Gates Foundation: Developing treatments for COVID-19 requires cross-sector collaboration:
Epidemics often bring huge challenges to the world. To protect people, especially the vulnerable, we need to find solutions to the problem - speeding up drug development and slowing the spread of the virus. The only way to treat viral infections like COVID-19 is with antiviral drugs. However, because this field is still blank, we can only treat the symptoms of different symptoms but not the root cause, and cannot treat various situations more thoroughly like antibiotics treat bacterial infections.
The following quote is from the WHO officially releasing the global research roadmap for COVID-19 (with Chinese version):
Figure 1-3 Research team, roadmap, knowledge gaps, etc., selected from the public health information public account
WHO officially releases global research roadmap for COVID-19 (with Chinese version)
Gates Foundation CEO: Developing treatments for COVID-19 requires cross-sector collaboration
IBM Summit The world's most powerful supercomputer is being used to fight the new coronavirus
DeepMind has launched a blockbuster weapon for predicting the "protein folding" of the new coronavirus: AlphaFold! Precision crushes biological and virus experts
Finding shortcuts to drug development: I'm no longer expensive
The father of YOLO announced his retirement from the CV industry and admitted that he could not ignore the negative impact of his work
Record the AI code of the 0302|Cell article and share it with those in China who are interested in using it.
The principle of closed loop optimization (CLO) system, pictures from the Internet (see the following news)
Machine learning helps electric vehicle battery development, and two-year charging test can be shortened to 16 days
https://tech.sina.com.cn/roll/2020-03-07/doc-iimxxstf7203224.shtml
Machine-translated from the Chinese original WHO研究路线图介绍及AI 潜在的应用场景 (published 2020-03-12). WeChat original ↗