"Preface: I have been a full 16 since I got my doctorate, but I still have a feeling in my PhD stage." I have been mentoring my first doctoral student in an independent laboratory for 13 years now, including 11 doctoral and postdoctoral students who have been pi in independent laboratories at universities in the United States and China. Their growth process is very different, and their personalities and abilities vary. It should be said that no one student can simply follow the footsteps of another good scientist to stand out. In this sense, the success of scientists is impossible to replicate. However, the common characteristics that good scientists often possess should be of great inspiration to young students. This article is mainly from my 2008 Tsinghua University postgraduate entrance education in a 2.5-hour lecture, but also a combination of some subsequent thinking and summary. In that lecture, I repeatedly emphasized that
My goal is not to ask graduate students to do exactly what I say, but hope that the fundamental impact, oscillation of the graduate students thinking angle, to inspire you to find the most suitable for their own success. This article is very long, divided into four parts successively published. 】
1. Time to pay. What all successful scientists must have in common is that they have to pay a lot of time and effort. This is a truth. In fact, no matter what kind of career in the community, to be the best in the industry, we must pay more than normal time. Sometimes, when an individual good scientist answers questions from students or the media, it is an understatement to say that success is a matter of luck, not toil. This answer is not objective and irresponsible, because they deliberately neglect their time to pay a lot, but only to emphasize a chance factor in the success process, so that the effect is often very misleading for young students, because some naïve students will even start to opportunistic, To wait for the so-called luck; other students begin to look for their luck and devote a considerable amount of their energy and time to things unrelated to scientific research. To be extreme: if there is a scientist who has a lot of luck rather than time to succeed, then his success is likely to be the result of grabbing someone else, and he doesn't have a truly leading academic level in the field. about a decade ago, a famous Chinese biologist, Mr. See, had a very well-known e-mail that was widely circulated on the Internet, and this email was written by Mr. Po to all of his PhD students and postdoctoral fellows, which I fully agree with. This email written in earnest, from which you can see Mr. Pu's hard intentions. I passed this email to all the students in my lab at Princeton and Tsinghua University to make them feel. One of the following is said: "The most important thing was what I consider to was sufficient amount of time and effort in the lab work. I mentioned that's about $ hr working time per week is what I consider the minimal time a average successful young Scienti St in these days have to put into the lab work ... I suggest that everyone puts in at least 6 HR concentrated bench work and A-H hr reading and other research-related Activit Y each day. Reading papers and books should be done Mostly after work. " (I think the most important thing is working time in the lab, and today a successful young scientist has an average of about 60 hours a week to get into the lab's research work ... I advise everyone to have at least 6 hours of intensive laboratory work and more than two hours of reading directly related to scientific research every day. Reading of literature and books should take place outside these working hours. )。 Some students read Mr. Po's email and told me, "it seems that I am not doing academic material, because I really can't eat this pain." "I often reply," When I was your age, I also felt that the long-term work was incredible. But unconsciously, you will gradually be touched by the subtlety of scientific research, will also be proud of their efforts and achievements, you will gradually adapt to this way of life! This sentence on the surface is to persuade students, but in fact is my own experience and experience. I have been particularly playful since childhood, do not like to study. But the education and pressure from the school and parents forced themselves to study hard as much as possible, I studied in Henan Province in the high School of Experimental Middle school, with more than others hard work, comprehensive results always ranked top. 1984 National High School Math League I won the first place in Henan division, walk into Tsinghua University. At the university level, I kept the tradition of hard work, combined with the first class and graduated one year in advance. But the result of this kind of test-taking and indoctrination is that I seldom really think independently and are not interested in the profession. When I graduated from college, I did not intend to engage in scientific research, but I wanted to do business in the sea. Between quirks, I embarked on the road to study in the United States. It is conceivable that the first year of study abroad, my mood swings very big, the inner impetuous and confused, the fundamental careless study, does the research, but spends a lot of time in the Chinese restaurant work, elective computer course. The next year, I began to gradually adapt to scientific research "boring", and began to have a bit of their own experience, sometimes understand some of the subtleties will be complacent, will also produce "the original so that" idea, gradually to their own scientific research ability has a little confidence. During this period, the doctoral graduate course all completed, I every week Five days, every day from 9 o'clock in the morning to do experiments to 7, 8, the weekend will go to two half a day. In the third year, I have begun to understand the logic of scientific research, a little eager to feel, in the group often ask questions, and this "getting Started" feeling let me add a lot of interest in research, at night often to do more than 11, to catch the last bus from Hopkins Medical School back to Homewood Campus (I live nearby). 1993 I used to mark the date of my experiment notebook "This is the 21st consecutive days of working in the lab." (This is my 21st consecutive day in the lab.) ) to motivate themselves. In fact, this is somewhat of a show of suspicion because itsOne of the weekends I've only done five or six hours of experimentation. By the year four, I have fully adapted to the laboratory's research environment, and will not feel the pressure of boredom or time. Schedule fully obey the needs of the experiment, as far as possible forward. In fact, this period of experimental time far more than just into the laboratory, but feel much better. In the late graduate stage, my hard work was famous in the laboratory. In New York, the post-doctoral period is the most bitter of my life two years, every night to do experiments to about three midnight, back to the place to lie down to sleep is often four points later, but every eight o'clock in the morning is the window of New York First Avenue on the noise of the car woke up, Back to the lab around nine o'clock and start a new day. Daily meals are in the laboratory, at 9 o'clock in the morning, 3 o'clock in the afternoon and 9, 10 o ' Night. This life rhythm lasts 11 days, from Monday to the second week of Friday, Friday night as Greyhound coach back to Baltimore (Baltimore) home, two days a weekend to sleep nearly 10 hours a day, to compensate for the past 11 days seriously missing sleep. Monday morning and start the next 11 days of struggle. Although physically tired, but my heart is very satisfied, very proud, I know that I am using action to build the future, in entrepreneurship. Sometimes I also encourage myself in my diary. I live near New York Manhattan District 65 Street and the first big road, close to the famous Central Park in New York, where there is also cultural entertainment, but for two years in New York, I have never taken a step in Central Park. I will certainly tell my own experience to each of my own students listen to, freshmen often ask me: "Teacher, you think you are bitter?" "I usually answer," It's hard to do things that you're not interested in. Be interested in the future not feel bitter at all. "Yes, a wonderful experiment brings me much better enjoyment than an American blockbuster. Now think back to the hard work at that time, feeling still very proud, very exciting! Sometimes I think: if the doctoral students, the post-doctoral stage of the 7.5 do not strive to forge ahead, but not to watch movies, read novels, to find entertainment (then the Internet is far from so rich in content), now what is the situation? to be a good doctoral student, time to pay is necessary.
2. The transformation of methodologyTo achieve breakthroughs and success in scientific research, only time and hard work is not enough. Critical Analysis (critical) is a must-have quality. The biggest difference between graduate students and undergraduates is that undergraduates learn from the accumulated knowledge of human beings, take into account scientific research and skills training, while the essence of doctoral students is to explore the creation of new knowledge through scientific research, and current and previous knowledge is to better serve the scientific research. In the undergraduate stage of learning knowledge-based, it is important to ask questions, but the answer is often there, so whether the question is critical is not so critical. Doctoral stage is completely different, must have the ability to critical analysis, otherwise it is impossible to become a good scientist. This is what I call a methodological shift. In fact, the essence of the whole university and postgraduate stage education is to cultivate the ability of critical analysis and develop the methodology of innovative scientific research. There are a lot of examples here, and the coverage is very wide, in a few examples that have made my life unforgettable.
(1) correct analysis of negative results (negative results) is the key to success. As a PhD student in the life sciences, if every experiment is successful and the expected positive results are achieved (positive results), in addition to the individual areas of study, it is generally only 6-24 months to obtain all the results needed for a doctorate. In fact, however, in the United States, a PhD student in the life sciences takes an average of about 6 years to get a PhD degree. This figure itself shows that the vast majority of experimental results are inconsistent with expectations, or negative results (negative results).Most junior doctoral students have negative views on negative outcomes, which directly affect their ability to develop critical analysis. In fact, as long as there is a proper control experiment (control experiments),
Correct judgmentThe negative results of the experiment are often the only road to success. In general, every exploratory subject has several or even more than 10 possible avenues of progress (hypothesis), the process of progress is basically to exclude the correct, to find the right direction of the process, in many cases, this is a few or even more than 10 possible ways to try, exclude, Until a workable path is found. In this process, a
trusted (conclusive)Negative results can often give us the confidence to abandon the current approach, and if used properly, this exclusion will ensure that we finally embark on the right experimental path. From this point of view, negative experimental results are not only normal, but also beneficial to the ultimate success of the project. Unfortunately, the negative results of most students are not convincing and cannot withstand logical scrutiny! This is often the biggest obstacle to the progress of scientific research projects. For example, according to a conventional protocol operation can not get the corresponding results of positive control, or the lack of a corresponding control experiment, or the credibility of the experimental results in the analysis and judgment of the error, so as to make "negative results" or "uncertainty" (Inconclusive results) conclusion, such a conclusion to the overall progress of the injury is very large, often let the students in the future experiment, overwhelmed and distressed. I warn and encourage all of my students: as long as you get the negative results of conclusive, your subject will soon be on the right track, and the strong logic analysis you have in the process of constantly analysing negative outcomes will also make you mature and grow into a good scientist. I am always worried about the smooth sailing and rarely negative results, because they have not really experienced the training of critical analysis in scientific research. In my lab, there are occasional students who have completed a PhD thesis in a very short time (about two years or even a year), and for these students, I'm sure to let them continue to take on some challenging new issues and get them tested for negative results. Without these disciplines, it is difficult for them to really have the ability to critical analysis and, in the future, to be an excellent scientist who can independently lead a laboratory. Sodon't be afraid of negative results, the key is to get the right information from the analysis of negative results.
(2) time-consuming perfectionism hinders innovation and progress. nikola Pavletich is my postdoctoral tutor and one of the most influential scientists, he has a strong sense of experimental and thinking ability, made a series including p53, Rb, CDK complex, SCF complex, BRCA1, a milestone in the field of structural biology, was a professor of president at the age of 31. In April 1996, I just arrived at the Nikola Lab shortly after, purified a very high expression of protein Smad4, two days down, the protein purification, but the results are not ideal: the yield may be only about 20% of the obtained. See Nikola, I am embarrassed to say: The yield is very low, I plan to continue to optimize the protein purification method, improve yield. He asked me: (to the effect) why does your want to improve the yield? Don ' t have a enough protein for crystallization trials? (Why do you want to increase the yield?) Is there enough protein for you to do the preliminary crystallization experiment? I replied: I do has enough protein for crystallization. But I need to optimize the yield first so I can accumulate more materials. (I have enough protein to do the crystallization screening, but I need to optimize the yield to get more protein.) He was rude to interrupt me: No. The yield is high enough. Your time is more important than yield. Proceed to crystallization. Wrong The yield is high enough, your time is more important than the yield. Please start crystallization as soon as possible. Practice has proved the value of Nikola recommendations. I crystallization screen with only a few milligrams of protein, and soon realized that this construct was not ideal and needed to remove dozens of amino acids from its N-terminal flexible by protein engineering. and the removal of N-terminal dozens of amino acid protein not only high expression, but also stable biochemical properties, purification is very easy, there is no need to worry about the yield problem. in the early stages of a bold innovation experiment, the design of each step of the experiment should be as careful as possible, but once the plan began to the intermediate steps of the experimental results do not have to pursue perfection, but should not hesitate to push the experiment step to the end, See if we can get an overall result that is broadly consistent with assumptions. Such asIn general, you should go back and carefully improve each step of the experimental design. If the overall design and operation of the experiment is not wrong, then your hypothesis (or overall direction) is likely to be a big problem. This methodology will be used in every day of the experiment. For example, in structural biology, the first attempt to purify a new protein should not pursue the yield of every step, but should try to carry out all the purification steps to the end, to see if they can get the protein suitable for crystallization. The first attempt to limited proteolysis, should not deliberately determine the protease concentration or the pursuit of protein purity, but to focus on whether the results have protease-resistant core domain. Since 1998, when I started my own independent laboratory, I told all the students: never pursue perfectionism. I push this methodology to the limit:as long as an experiment can go forward, we must achieve the end, as far as possible to see the results of each step, after the need to look back, one by one to solve the problems encountered in the middle.
(3) Research literature (literature) and academic lectures (seminar). nikola Pavletich is knowledgeable. In the minds of many of our postdoctoral fellows, he must have read many articles and often listened to seminar. Unexpectedly, my greatest astonishment appeared in my conviction of this point. at my doctoral stage, my mentor Jeremy Berg attaches great importance to reading relevant scientific literature, and has weekly Journal club to discuss important research progress. Just arrived at the Nikola Laboratory, I tried to show myself reading paper's foundation, also want to discuss with Nikola at the same time to get his true line. One day in the spring of 96, I read a "Nature" article, met Nikola before lunch to describe the subtlety of the article and look forward to his comments. Nikola looked a little embarrassed to say to me: "Sorry, I have not read this article." Oh, maybe this article is too new and he hasn't had time to read it. After a few days, I read a few months ago published the "Science" study article, and to find Nikola discussion, did not think he said he had not seen. After several snags, I puzzled to ask nikola:you know so much. You must read a lot of papers. Why are it that you didn ' t read the ones I read? (your knowledge is so profound, you must have read a lot of literature extensively.) Why didn't you just read the papers I mentioned? Nikola looked at me and said: I don ' t read a lot. (I don't read extensively.) I ask: If you don ' t read a lot, how can I good at? And how can I reference so many papers in your own publications? (if you don't read extensively, how can your research be so good?) How can you cite so many documents in your own paper? Nikola's answer made me completely surprised: (to the effect) I only read papers that is directly relevant to my the interests, and I only read more P Apers when it comes to writing my own papers (I only read papers that have a direct relationship to my research interests. And I read a lot only when I'm writing a paper. ) I do DoctorAfter the unit Memorial sloan-kettering Cancer Center has a good series of academic lectures (Institute Seminar series), often invited to various life sciences in the field of Daniel to speak. On one occasion, a Nobel Laureate Institute seminar, and named Nikola to talk to him. For the vast majority of people, this is a rare opportunity to get close to the big shots and get a good impression. Nikola told his secretary: please convey my apologies for me, seminar I happened to be absent that day. We also regret for Nikola. To my surprise, the Nobel Laureate Seminar's Day, Nikola locked himself in the office, in the morning until late in the evening has not gone out, of course, not to listen to lectures. Of course, this may be a coincidence –nikola canceled his trip plan, but with our understanding of Nikola, he is in all likelihood is writing paper. Later, we also realized that this kind of thing happened in the Nikola body has seen how not to blame. before I left the Nikola Lab, I asked him, "How can you be such a good scientist if you don't read papers and don't listen to lectures?" He replied: My time is limited, only about 10 hours a day in the laboratory, after weighing the pros and cons, I can only use my limited time in what I think is the most important things, such as analytic structure, analysis of the structure, and students to discuss issues, writing articles. If there is not enough time, I can only read less articles and listen to lectures less. nikola's answer expresses a simple truth: a person must make a choice about what he does, and it cannot be exhaustive. Whether it is the reading of scientific literature or the listening of academic lectures, it is to learn from relevant experience and to better serve their scientific research projects. in the doctoral stage, especially the first two years, I think it is necessary to spend enough time to listen to the academic lectures in the relevant fields, and to conduct extensive reading of the scientific literature, and to lay a good foundation for critical thinking; but with the deepening of scientific research, the literature reading and academic lectures need some pertinence, It is also time to start weighing the allocation.
(4) challenging traditional thinking From my sensible beginning, have been educated: all things fail to have its reason, should find the cause of failure and then start again. It was not until 1996 that I followed this principle in my experiments. But in Nikola's laboratory, this basic principle is also subject to a well-reasoned challenge. one time, a more complicated experiment failed. I was frustrated and prepared to spend a few days doing more control experiments to find the problem. Unexpectedly, Nikola stopped me, and he frowned and asked me, (to the effect) tell me why do you want to figure out why your experiment failed? (Tell me why do you understand why the experiment failed?) I think this question is too unreasonable, righteously answer: I need to know what went wrong so I can get it to work next time. (I need to know what is wrong to guarantee the next time can be successful.) Nikola immediately commented: (to the effect) you don ' t need to. All carefully repeat your experiment and hopefully it'll work next time need. Many times figuring out why your previous experiment failed would take much longer time than simply repeating your experime Nt. For a sophisticated, one-time experiment, the best solution to a failed experiment are to repeat it carefully. (not required.) What you really want to do is to repeat the experiment, maybe next time you can do it. Instead of spending a lot of time figuring out why an experiment failed, it's better to repeat it first. The best way to face a failed, complex, one-off experiment is to do it again. Later, Nikola his point of view: (to the effect) It is a philosophical decision whether to the figure out what an experiment failed. The conventional wisdom of understanding EVery glitch may not be represent the best approach. Think about it, these words make sense. Not all failed experiments must find its cause, especially the life science experiments, the process is cumbersome and complex; Most failed experiments are caused by simple manipulation errors, such as PCR forgetting to add a certain ingredient, which can be done over and over again, often solving the problem. Only those experiments that are critical and fail to find the cause of failure need root planing. The examples I have chosen are somewhat "extreme", but only in this way can we better play a role in shaking people's minds. In fact, in my own laboratory, these examples have been repeated over and over again to all the students, and each time I finish, I will tell you to break the superstition, suspicion of stereotypes, and the key is: follow logic! (Follow the logic!) I was destined to repeat in the lab every day: follow logic! Talk to different students every day, plus at least 5 times more. And my own every time with doctoral students discuss the subject also always follow the strict logic, use reasoning, elimination method to find the next step of the solution of the experiment. strict logic is the root of critical analysis。 from:http://blog.sciencenet.cn/home.php?mod=space&uid=46212&do=blog&id=484416http:// Blog.sciencenet.cn/blog-46212-486270.html
How to be a good doctoral student--Professor Shi
