Computational Biology/Bioinformatics in PhD Portion of MD/PhD

[thefiddler]

Hello. I've been strongly considering getting an MD/PhD. My current interests are in computational biology/bioinformatics, and image analysis/illness representations. I consider Radiology and Immunology to be particularly strong specializations to augment my research, which would mostly be computational. I am not currently asking whether an MD/PhD is worth it for this particular set of interests, though I would appreciate any input in that regard. My true question is what a PhD in computational biology/bioinformatics would consist of in scheme of the larger MD/PhD. Is the PhD experience generally very similar for PhD-only students and MD/PhD students? Or would my PhD also be as translational as I would like following the completion of the dual degree?

Thank you very much.

---

Members don't see this ad.

 

[Neuronix]

Is the PhD experience generally very similar for PhD-only students and MD/PhD students?

Yes.

would my PhD also be as translational as I would like following the completion of the dual degree?

The idea is to train in and produce fundamental work for a later career including translational work.

 

[thefiddler]

Thank you for your feedback, Neuronix. If I may ask a follow-up question. Many computational biology programs have graduate classes to teach students the necessary skills to do their PhD work. This makes the programs approximately 6 years long. Should I expect my PhD portion to also be as long if the experience is mostly the same, or have you heard that MD/PhD students have an altered curriculum to shorten it in order to maintain a 7-8 year MD/PhD?

 

[eteshoe]

Thank you for your feedback, Neuronix. If I may ask a follow-up question. Many computational biology programs have graduate classes to teach students the necessary skills to do their PhD work. This makes the programs approximately 6 years long. Should I expect my PhD portion to also be as long if the experience is mostly the same, or have you heard that MD/PhD students have an altered curriculum to shorten it in order to maintain a 7-8 year MD/PhD?

If you have no coding experience, you may need to take the necessary courses your first yr of the PhD. For many comp bio programs, a lot of the students already have coding experience, and the classes usually help put it in the context of biology (since they tend to get a lot of math and CS applicants). If you want to minimize your PhD time, it would be better to join some program like mol bio, immunology, biophysics, genomics, etc - find a PI that has comp bio projects in the lab, and take the appropriate classes as electives (@Neuronix may have a better take on the situation) to supplement your deficiencies. Or you can do what I'm doing - find a PI who has comp bio projects, and spend a good deal of time teaching yourself (and bugging your CS/comp bio friends) whatever coding language is needed. Plus Google.

Also the avg for the dual degree program is ~8 (I know a good deal of 9 yrs as well, and the occasional 10 yr student), not that many people finish w/ a 3 yr PhD these days.

 

[Neuronix]

or have you heard that MD/PhD students have an altered curriculum to shorten it in order to maintain a 7-8 year MD/PhD?

I hear all kinds of things... I also completed an MD/PhD program and got a few t-shirts.


These are pretty basic questions you're asking. It might help your understanding to do some reading of the official materials.

https://students-residents.aamc.org...ticle/education-and-training-md-phd-programs/

"Since the average time to complete a biomedical PhD in the U.S. is about six years, by integrating the didactic components of training, dual degree training may require less time to complete than if each degree were pursued independently."


As eteshoe points out, the average for MD/PhD programs is actually about 8 years. Therefore, mathematically (and in reality), the programs are 7-9 years long.

With the trends upwards I was seeing years ago, I wouldn't be surprised if the average was more like 8.5 nowadays. At some programs I've been told that it is.

 

[globe199]

If you have no coding experience, you may need to take the necessary courses your first yr of the PhD. For many comp bio programs, a lot of the students already have coding experience, and the classes usually help put it in the context of biology (since they tend to get a lot of math and CS applicants). If you want to minimize your PhD time, it would be better to join some program like mol bio, immunology, biophysics, genomics, etc - find a PI that has comp bio projects in the lab, and take the appropriate classes as electives (@Neuronix may have a better take on the situation) to supplement your deficiencies. Or you can do what I'm doing - find a PI who has comp bio projects, and spend a good deal of time teaching yourself (and bugging your CS/comp bio friends) whatever coding language is needed. Plus Google.

Also the avg for the dual degree program is ~8 (I know a good deal of 9 yrs as well, and the occasional 10 yr student), not that many people finish w/ a 3 yr PhD these days.

I actually joined a computational biology lab for my PhD without any prior coding experience - just a passion for physics and math. I also only rotated in labs in the computational biology graduate program! I always kept telling myself that I'd teach myself to code while in the preclinical curriculum, but I found that there was just no time if I wanted to keep myself on top of all the material. By the time I actually started my PhD, I barely remembered any coding I learned during my lab rotations or first-year computational biology course. I was glad I at least remembered how to navigate a shell window!

I ended up just "shadowing" a postdoc in the lab I joined for the first week or so to get familiar with working on a computational project again. I found myself relying on the postdoc or the internet (mostly StackOverflow) whenever I needed to find out how to do something. It took a couple of months before I was comfortable with not looking something up every few minutes.

I've actually often heard that a computational biology MD/PhD is faster than a wet-lab one. Anecdotally, from the handful of students I've known, 7-year MD/PhDs are more common for computational projects. Like for anyone else, though, at the end of the day it depends on the whims of one's thesis committee!

 

[thefiddler]

I see, thank you all very much.
I have a significant amount of coding experience, albeit incomplete: C++, Java, Python, Assembly...But my current research experience is somewhat limited to basic science/wet lab work. I have begun collaborating with a postdoc in my lab to write a program to automate the image analysis that needs to be done, but it isn't nearly complex enough for me to extrapolate.
I do recognize that my questions are rather basic; it is mostly because I do not have much exposure to computation-based projects and how long they take. If I was interested in a traditional genetics or molecular biology PhD, I have a pretty solid idea of how much time and effort 3-4 published papers would take.
For those of you who have completed or are doing a computation biology-based PhD: do you find having the 2 years of didactic medical school classes (and the MD as a whole) facilitate how you program your algorithms or organize your code to answer questions which have a disease basis?
The reason for all of my questions is because I wonder if having a prolonged, more thorough education (presumably the dual degree) would prepare me best for a career in computational medicine, or if the extra four or so years would be more wisely spent gaining experience working alongside doctors as a computational biologist (the PhD-only; with the caveat that I may not be able to work on disease models that are not currently being studied). Of course, I know that with my prior coding experience, getting an MD and then doing research using computational biology techniques is also a possibility, though I'm not yet sure if that is a route I want to consider currently as my end goal involves spending the majority of my time doing research.

 

[globe199]

I see, thank you all very much.
I have a significant amount of coding experience, albeit incomplete: C++, Java, Python, Assembly...But my current research experience is somewhat limited to basic science/wet lab work. I have begun collaborating with a postdoc in my lab to write a program to automate the image analysis that needs to be done, but it isn't nearly complex enough for me to extrapolate.
I do recognize that my questions are rather basic; it is mostly because I do not have much exposure to computation-based projects and how long they take. If I was interested in a traditional genetics or molecular biology PhD, I have a pretty solid idea of how much time and effort 3-4 published papers would take.
For those of you who have completed or are doing a computation biology-based PhD: do you find having the 2 years of didactic medical school classes (and the MD as a whole) facilitate how you program your algorithms or organize your code to answer questions which have a disease basis?
The reason for all of my questions is because I wonder if having a prolonged, more thorough education (presumably the dual degree) would prepare me best for a career in computational medicine, or if the extra four or so years would be more wisely spent gaining experience working alongside doctors as a computational biologist (the PhD-only; with the caveat that I may not be able to work on disease models that are not currently being studied). Of course, I know that with my prior coding experience, getting an MD and then doing research using computational biology techniques is also a possibility, though I'm not yet sure if that is a route I want to consider currently as my end goal involves spending the majority of my time doing research.

I personally found the MD curriculum to be helpful in contextualizing much of the biomedical research I've come across - it's made it much easier to be familiar with the biological background knowledge. It's also made it easier to understand where some papers and seminars are coming from. Not so for actual programming or organizing code, though, since a traditional medical school curriculum doesn't teach that. It's not like medical school would help much with a wet-lab PhD, either: it isn't going to teach you how to actually do X-ray crystallography or Cre-Lox recombination, for example.

In biomedical research, at the end of the day both programming and experiments are means to an end. While MD training may not help you learn the means, it can help you understand that end. Only a PhD or other extensive research training will help you learn the means. Therein lies the value of MD/PhD training!

Also, even if you want to be a majority researcher, try to make sure that the MD training is something you could see yourself actually enjoying before starting an MD/PhD program. Medical school is a lot of work, and if you don't enjoy it it will only feel like wasted time. Painful, wasted time...

 

[thefiddler]

I really appreciate your perspective, globe199.

I completely understand what you mean regarding the MD training. I have wanted to be a physician for about eight years now, but my perspective on how best to help my patients has since developed greatly. I do believe I will enjoy the training, I just also have very defined goals which I would like to meet, and if the MD-only option is not the most conducive to my meeting those goals, I am happier taking a longer and, in different ways, more difficult path.

It is also pleasing to read that the MD is helpful for contextualizing the research, which is what I was trying to ask. I find my programs are more optimized when I have a better grasp of the subject it is being applied to, but I wasn't sure if the same applied with an MD/PhD in computational biology.

Thank you for this line, in particular: "In biomedical research, at the end of the day both programming and experiments are means to an end. While MD training may not help you learn the means, it can help you understand that end."

 

[ellealla]

I am doing an MD and a PhD in computer science. It's a very useful combination. If you do a computational PhD you will likely have to take many more courses than your classmates doing biological PhDs. For example, some of my classmates have to take zero courses. I have to take six. Because there are a couple grad courses in statistics I want to take for my research, I'll probably end up taking 8 courses. But, you can learn a lot through courses, so it's worth it to me. There are a lot of opportunities at the intersection of quantitative fields and medicine. I'd say go for it!