Traits of Successful Grad Students

[tortuga87]

---

Members do not see this ad.

Hi everyone,

As I am looking through faculty web pages, occasionally I come across these amazing grad students that churn out 2-3 quality first author papers a year. One guy I saw almost seems like he's going to get one professor tenure single handedly! Many of the other grad students (even in the same lab) only manage about 1-2 first author papers during their entire grad school years.

Does anyone know what's the difference between these two groups? Does the amazing group develop an extremely clear game plan before they start? Do they not sleep? I can imagine some factors like working very well with your mentor and lab, but I'm wondering if I can get some more specific advice. Because I would like to be as successful too of course, if possible 😛

I'm also curious if such productivity is more a function of your own hard work or which lab you choose, or how they are related


Similar thread about MD/PhD in general:

http://forums.studentdoctor.net/showthread.php?t=725387

Thanks!

 

[wuhanese]

http://www.technologyreview.com/blog/boyden/21925/#comments

nothing magical really - hard work, intelligence, focus, efficiency, and consistency. you have to breathe science when you sleep

 

[zelbest]

Besides the traits wuhanese mentioned, quite a bit of luck is involved in science. The best you can do is to make sure you have a good understanding of the field before you start, or else you might spend a year chasing ghosts.

 

[Shifty B]

This is a really interesting question. I had, by most standards, a very productive PhD and will share some advice that helped me. You can PM me or post here if you have specific questions.

I have to start out by saying that there is a huge component of luck involved. It is also very lab and advisor specific. Some fields are more prone to rapid publication of small developments while other fields you just publish one Nature paper with an entire thesis worth of work. Most of this advice applies either way. I'm going to leave out any advice about how to choose an advisor.

I'll also try to focus on very specific advice. Of course you should think about science all the time, put in the hours, read papers, blah blah blah. That advice isn't useful because you already know that.

So:
1) Work on multiple projects at once. You know that down time you have between gels, or while you are waiting for your mice to breed. Work on something else during that time. Get 1 or more side projects. Odds are, after time, your side project will end up being your main project after your main project bombs. At any given time you should be doing 3 or 4 different things.

2) Collaborate with as many people as possible. Once you've been in a lab for 1-2 years, you probably have developed 1 or more skills or some knowledge set that no one else at your institution has. Talk to other people in your department and other departments and think of ways you can apply your skill to their niche. Every conference you go to, think about how it applies to your work. Then start a new side project based on your ideas (see #1 above).

3) Start writing the paper as soon as you start the project. Literally. Sit down and write an outline of what you expect your paper to be. Plan your introduction, write the methods, plan the sections of the results. Prepare a list of what you expect each individual figure to be. Then, start doing the work that you need to obtain that data. I can't overemphasize this. It's so much easier to decide what to do when you've already planned it out. It also makes #4 much easier.

4) Limit your unnecessary experiments. If it doesn't pertain to your plan for your paper (or at least a stepwise way to get toward it), don't do it. If I had a nickle for every time I saw someone do an experiment for a week that they didn't need to do because the results were irrelevant to their final message, I'd be rich. You might as well come in, pour reagents down the drain, and go home.

5) Split huge papers into 2 managable papers. I'm sure someone is going to take exception to this, but whatever. Also, some fields don't tolerate this. However, a paper doesn't need to have 8 figures with 4 panels each. The best papers I ever read have a single, manageable message, that is made simply. I put it down and I can summarize in 1 or 2 sentences what it means. Think about that when you're doing your work.

6) Don't get tripped up by personality conflicts. Navigating grad school is like overseeing negotiations between Israel and Palestine. You need to be freakin Madeleine Albright out there, working both sides. You have to stand up to your advisor, your committee members, and everyone you work with by sheer force of will and bend them in the direction that helps you the most. You need allies other than your advisor, and I recommend you make friends with the biggest heavyweight in your department. Put him/her on your committee, and convince him you are right. Everyone else will fall in line.

7) Be willing to accept publication in a lesser journal, sometimes. Before you run that western blot for the 100th straight time on the off chance that Cell is going to accept your paper, take a realistic look in the mirror and assess your chances. It may be better to get your paper published in a well read specialty journal and just move on to your next thing. If you're in danger of getting scooped especially, maybe just get the paper out there to lay the groundwork for the one that does have a chance. This is a delicate balance, because you want to publish in the best available journals. However, spending a year working on some bull**** to get in a journal that has an impact factor of 9 instead of 6, forget it.

This is only a start, but maybe it will help someone.

 

[soundnin]

This is a really interesting question. I had, by most standards, a very productive PhD and will share some advice that helped me. You can PM me or post here if you have specific questions.

I have to start out by saying that there is a huge component of luck involved. It is also very lab and advisor specific. Some fields are more prone to rapid publication of small developments while other fields you just publish one Nature paper with an entire thesis worth of work. Most of this advice applies either way. I'm going to leave out any advice about how to choose an advisor.

I'll also try to focus on very specific advice. Of course you should think about science all the time, put in the hours, read papers, blah blah blah. That advice isn't useful because you already know that.

So:
1) Work on multiple projects at once. You know that down time you have between gels, or while you are waiting for your mice to breed. Work on something else during that time. Get 1 or more side projects. Odds are, after time, your side project will end up being your main project after your main project bombs. At any given time you should be doing 3 or 4 different things.

2) Collaborate with as many people as possible. Once you've been in a lab for 1-2 years, you probably have developed 1 or more skills or some knowledge set that no one else at your institution has. Talk to other people in your department and other departments and think of ways you can apply your skill to their niche. Every conference you go to, think about how it applies to your work. Then start a new side project based on your ideas (see #1 above).

3) Start writing the paper as soon as you start the project. Literally. Sit down and write an outline of what you expect your paper to be. Plan your introduction, write the methods, plan the sections of the results. Prepare a list of what you expect each individual figure to be. Then, start doing the work that you need to obtain that data. I can't overemphasize this. It's so much easier to decide what to do when you've already planned it out. It also makes #4 much easier.

4) Limit your unnecessary experiments. If it doesn't pertain to your plan for your paper (or at least a stepwise way to get toward it), don't do it. If I had a nickle for every time I saw someone do an experiment for a week that they didn't need to do because the results were irrelevant to their final message, I'd be rich. You might as well come in, pour reagents down the drain, and go home.

5) Split huge papers into 2 managable papers. I'm sure someone is going to take exception to this, but whatever. Also, some fields don't tolerate this. However, a paper doesn't need to have 8 figures with 4 panels each. The best papers I ever read have a single, manageable message, that is made simply. I put it down and I can summarize in 1 or 2 sentences what it means. Think about that when you're doing your work.

6) Don't get tripped up by personality conflicts. Navigating grad school is like overseeing negotiations between Israel and Palestine. You need to be freakin Madeleine Albright out there, working both sides. You have to stand up to your advisor, your committee members, and everyone you work with by sheer force of will and bend them in the direction that helps you the most. You need allies other than your advisor, and I recommend you make friends with the biggest heavyweight in your department. Put him/her on your committee, and convince him you are right. Everyone else will fall in line.

7) Be willing to accept publication in a lesser journal, sometimes. Before you run that western blot for the 100th straight time on the off chance that Cell is going to accept your paper, take a realistic look in the mirror and assess your chances. It may be better to get your paper published in a well read specialty journal and just move on to your next thing. If you're in danger of getting scooped especially, maybe just get the paper out there to lay the groundwork for the one that does have a chance. This is a delicate balance, because you want to publish in the best available journals. However, spending a year working on some bull**** to get in a journal that has an impact factor of 9 instead of 6, forget it.

This is only a start, but maybe it will help someone.

wow great advice. the PI i worked for tried to instill these things into me as well (except for #5 and 7 since he was a new faculty member , which did result in us being scooped once...). i totally agree with all of that, especially the idea of never wasting your time. you set out with an outline of your paper in mind, design all your experiments so the data could potentially end up in a figure panel, and in your spare time, work on side projects. truly well put solid advice. 👍

 

[Lil Mick]

Here are a few suggestions that are working well for me:
1) Multiple projects
2) Collaboration
3) Submit to journals widely read by your intended audience (maybe not Cell or Nature)
4) Projects with lots of data collected already (math/epi/engineering)
5) Time management (if you're still in MS1 or MS2)

 

[BarmyArmy]

Always remember to design your project around a robust system. It will save you years of frustration.

 

[Shifty B]

4) Projects with lots of data collected already (math/epi/engineering)

So, I forgot to mention this, but it can be huge as Lil Mick pointed out. If you have some stack of data already collected, see what else you can do with it. Can you reprocess it some way to get some new findings? Can you come up with a mathematical model to describe it? If so, it can possibly become a new paper.

Also valuable is work that the lab did already but someone left behind. Not infrequently, a grad student or postdoc will finish their stay before quite finishing some loose ends. When starting as a grad student, it can be useful to pick up these projects and see what can be done to make them work, even if your project is already something else.

 

[Shifty B]

Always remember to design your project around a robust system. It will save you years of frustration.

Well, what do you mean by "robust system"?

I suppose it could mean an experimental design and setup that was likely to work in a variety of situations and continue to withstand challenges and failures. This tip is true but it's not very likely to be useful. In most cases it's impossible to tell if this is true when you start, and you won't find out until you crash and burn later.

 

[BarmyArmy]

Well, what do you mean by "robust system"?

I suppose it could mean an experimental design and setup that was likely to work in a variety of situations and continue to withstand challenges and failures. This tip is true but it's not very likely to be useful. In most cases it's impossible to tell if this is true when you start, and you won't find out until you crash and burn later.

Its exactly that. I agree it is hard for student to recognize that in the beginning but a good sign is that if the postdocs or more senior grad students have been having problems with the system consistently, then its probably a good idea to shift away from it.

 

[tortuga87]

Thanks for all the great advice everyone

Well, what do you mean by "robust system"?

I suppose it could mean an experimental design and setup that was likely to work in a variety of situations and continue to withstand challenges and failures. This tip is true but it's not very likely to be useful. In most cases it's impossible to tell if this is true when you start, and you won't find out until you crash and burn later.

The robust system is an interesting idea... haven't thought of that.... I'm guessing this also implies that, if you do decide to work with a new PI, you should go to a lab that has been around for at least a few years and has postdocs as opposed to trying to be the first star student. Starting in an entirely new lab would probably also mean you would have to focus on the nitty gritty technical details rather than the science. This is kind of a problem for me, since I am in a relatively new field, so lots of assistant professors

So far my equation has been to include all advice mentioned above, focus on labs that had created a few star students in the past (so that I know there is at least a possibility), and published a lot with only a few lab members. Though this doesn't leave many labs hmm...

 

[tortuga87]

Some other advice from my mentor to add to the collection:

1) Focus on one thing and be really good at it. I.e., There are many interesting things, but don't try to be the Renaissance man by frequently changing your research interests and shifting your research topics in grad school/residency/fellowship. Faculty recruiting committees generally want a solid research interest with a proven publication record

2) Make an effort to always talk to people and express gratitude in person. If you talk via email or some other way, relationships tend to grow distant overtime and bad things may happen because of that