Both are known as the "Nobel Prize weathervanes", but who is the most reliable? Where does the charm of the Nobel Prize come from?

On September 30, as the 2023 Nobel Prize is about to be announced, "Fanpu" invited Dr. Xu Yixun, an author of this journal and a senior R&D expert in the field of biopharmaceuticals, to make a live broadcast of his predictions for this year's Nobel Prize in Physiology or Medicine (the full content of the live broadcast can be seen on the Fanpu WeChat video account). In the Q&A session, the host asked several representative questions on behalf of the audience. These questions can be summed up in a popular saying, "Why has the Nobel Prize always been a model and never been surpassed?" So many awards use the name of the Nobel Prize to establish themselves and promote themselves. Who can get the most out of it?

The following text has been reorganized based on the live Q&A for reference.

Speaker | Xu Yixun

Host | Ding Zong

With so many Nobel Prize indicators, which one is the most reliable?

Q: There are always some awards that are called "Nobel Prize weathervanes", such as the Wolf Prize in chemistry and the Lasker Prize in biology. Are the selection criteria of these awards similar to those of the Nobel Prize?

Xu: The predictive performance of different awards as a weathervane for the Nobel Prize varies greatly, and changes with time. According to past historical records, the Wolf Prize in Chemistry and the Prize in Medicine are not very predictive of the corresponding Nobel Prizes. Sometimes, even if the work of the awardees matches, the arrangement of candidates is relatively less fair than that of the Nobel Prize. For the Nobel Prize in Physiology or Medicine, the Lasker Basic Medical Science Award has performed very well as a "Nobel Prize weathervane" since its establishment in 1946 until around 2009. Unfortunately, since 2010, the Lasker Award and the Nobel Prize's judging criteria have begun to drift apart. The Horwitz Award awarded by Columbia University in the United States, although lacking in popularity due to lack of media attention, from 1967 to 2014, more than 50% of the winners later won the Nobel Prize in Biomedicine or Chemistry! And many of the important works that hit the Nobel Prize in Biomedicine have never been favored by the Lasker Award. Unfortunately, no Horwitz Award winner since 2015 has won the Nobel Prize. In general, there are not many awards that have the predictive power of a "Nobel Prize weathervane", and only the Horwitz Prize and the Lasker Prize stand out.

Q: So is it feasible for someone to use AI tools such as ChatGPT to predict the Nobel Prize this year?

Xu: I don’t think it’s reliable. The way ChatGPT, based on a large language model, predicts the Nobel Prize is similar to the so-called “citation award” that Thomson Reuters throws out every year on the Internet. That is, by simply counting the number of citations of all the papers published by each scientific award winner, it is believed that the top ones are strong contenders for the Nobel Prize. This scoring standard for evaluating scientific research results is obviously too simple. Some papers with average scientific research value may also receive high citations due to some special reasons, so the correlation between the number of citations and the recognized value of academic papers is weak. The data used to train ChatGPT is public language text from the Internet. The large language model obviously cannot obtain peer-reviewed value judgments on scientific discoveries or inventions. Thomson Reuters’ prediction power for the Nobel Prize in the past few decades is far lower than that of the Horwitz Prize and the Lasker Prize, and I think ChatGPT’s prediction performance will be similar to the “citation award”. In the short term, I cannot see how the AI ​​large language model can estimate the technical content of each scientific research result.

Why has the Nobel Prize always been imitated but never surpassed?

Q: Is the Nobel Prize a pioneering award? Or should it be a pioneering award?

Xu: My answer to both questions is yes. A similar question was raised by an audience member during last year's live lecture. The leadership of the Nobel Prize is mainly reflected in the allocation of scientific research funds by government management departments of various countries. How to ensure fairness in the allocation of funds for different disciplines and branches is a very difficult task. Life sciences are diverse, and researchers in each field believe that their research direction is more important. The Nobel Prize science history of more than 120 years can give management departments and scientific research fund expert review committees a relatively objective value orientation. For many basic research projects, it is likely that their future importance cannot be predicted when reviewing their fund application bids. At this time, if the reviewers can learn from the "big data" of the Nobel Prize science history and fund some seemingly "high-risk" topics with an open mind, they may see that those basic research driven by curiosity have unexpectedly made important application breakthroughs . For example, the history of the discovery of green fluorescent protein that I have published (see "It is difficult to open a road without a road: the legendary discovery journey of green fluorescent protein"). In contrast, the “War on Cancer” initiated by former US President Nixon in 1971, despite the huge investment in scientific research funds, had little effect until the end of the 20th century. The immune checkpoint inhibitors that won the 2018 Nobel Prize in Medicine were the result of the discovery of the PD-1/PD-L1 pathway that regulates T cell immunity by basic immunologists, which brought a milestone breakthrough in cancer treatment in an unexpected way that could not be predicted in advance.

Q: This inevitably reminds me of the good book "Why Greatness Cannot Be Planned" published by Springer in 2015, which may also express a similar idea. The leadership of the Nobel Prize does not seem to be the leadership of a specific goal, but because of the breakthrough in the basic principles of a field, and the macro leadership in this direction.

Xu: That’s right! And it makes sense that the Nobel Prize has always attached more importance to basic research. Good basic research has no application prospects at the beginning. This is actually reflected in today’s live lecture on the history of the discovery of UPR. Peter Walter had sufficient funding to study yeast at the University of California, San Francisco, and no administrative staff ever asked him to pay attention to the application or transformation of research results. After leaving the United States and returning to Japan, Mori Kazutoshi joined the Heat Shock Protein Institute (HSP Institute), and his research has always been interfered with by administrative staff in the dominant direction. This meant that although he barely maintained his independence from Walter’s laboratory in important discoveries through the good foundation laid by the scientific training he received in the United States and hard work, he was left far behind in the depth and quality of his scientific research results. In this case, the scientific research management departments in the United States are obviously better than those in Japan in terms of macro-guidance in line with the Nobel Prize history of science.

Q: Is there any relationship between teachers and students in Nobel Prizes? Because we mentioned before that there are many kinships between laboratories, this person goes there, that person comes here. Also, in physics, for example, the Curies won the prize for two generations. Is there a similar situation in the fields of physiology and medicine?

Xu: The probability of winning the Nobel Prize does have the characteristics of inheritance from kinship and mentorship. The Curies won the prize successively, which obviously has the component of high intelligence inherited through genes. In addition, since parents are role models for their children, being influenced by their parents' Nobel Prize-level scientific contributions will have a positive impact on their children's education, growth and career planning. For example, Arthur Kornberg shared the Physiology Prize with Severo Ochoa in 1959 through his research on DNA polymerase. After his son Roger Kornberg became an independent scientist, he chose to study RNA polymerase and the regulatory mechanism of transcription, and won the Chemistry Prize in 2006. Because of his father's academic prestige, Roger was able to go to a first-class laboratory to conduct research during his undergraduate years, and it was easy for him to have the opportunity to publish papers and obtain many resources that ordinary undergraduates could not reach.

In addition, the mentor-apprentice relationship between the mentor and the doctoral students and postdoctoral fellows is more important than the kinship relationship. Peter Walter, who is mentioned in today's live lecture and may win the award, has a doctoral thesis supervisor who is Gunter Blobel, the winner of the 1999 Nobel Prize in Physiology. He has made important discoveries and published first-class papers before getting his degree. To be more realistic, Walter made important discoveries about the unfolded protein response (UPR) after becoming independent. As a Nobel Prize winner, Blobel can continue to nominate him to the Nobel Committee. In addition, Nobel Prize winners can teach graduate students and postdoctoral fellows by example in terms of their vision in choosing research topics and their ability to ask important questions. This is vividly described in Robert Kanigel's 1986 masterpiece "Teaching Genius".

This article is supported by the Science Popularization China Starry Sky Project

Produced by: China Association for Science and Technology Department of Science Popularization

Producer: China Science and Technology Press Co., Ltd., Beijing Zhongke Xinghe Culture Media Co., Ltd.

Special Tips

1. Go to the "Featured Column" at the bottom of the menu of the "Fanpu" WeChat public account to read a series of popular science articles on different topics.

2. Fanpu provides a function to search articles by month. Follow the official account and reply with the four-digit year + month, such as "1903", to get the article index for March 2019, and so on.

Copyright statement: Personal forwarding is welcome. Any form of media or organization is not allowed to reprint or excerpt without authorization. For reprint authorization, please contact the backstage of the "Fanpu" WeChat public account.