3 ms·

As a physician who was a mathematics major in undergrad, I completely agree. An astonishing amount of biology is rather mechanistic, systematic, and logical. I

As a physician who was a mathematics major in undergrad, I completely agree. An astonishing amount of biology is rather mechanistic, systematic, and logical. I feel my math (and minor programming) experience trained my brain in such a way as to understand complex biological interactions more intuitively. I'm not bragging here, but while I was actually understanding the theories behind what we were learning, on a fundamental systemic basis, many of my classmates were essentially relying on memorization. They passed the exams fine, and make okay doctors, but they can't explain things to patients well and their research ideas are very limited and safe.

maxerickson 10y ago
But how did your biology undergrad go?
- maxerickson 10y ago
  Right, I forgot that sarcasm is the worst form of sarcasm. The GP is implicitly comparing the impact a math curriculum had on their reasoning with the impact that a biology curriculum might have had. But n=1 for the math curriculum (perhaps the beneficial traits are endogenous) and n=0 for the biology curriculum.
qud 10y ago
How did you transition from a Maths/Programming undergrad to doing medicine? I'm doing Computer engineering and plan to do go into medicine later, any tips?
- evoloution 10y ago
  If I were you: 1) I would go work in a computational biology lab, get some papers out, Cancer research is always a step ahead but there a lot of good projects in Neuroscience too (immunology could also work). Go to the best lab you have around, most PIs know that it is hard to find good programmers because they mostly go to the big tech companies or startups. Make sure you don't do wet lab at this step it is going to be a waste of your time since you will not have time to do anything meaningful. 2) With the papers apply to medicine or preferably to MD/PhD program, if you get into MD/PhD program you will have time to learn some real science too. In any case if you are the ambitious type of person when doing step 1 try not to get stuck working as a bioinformatician: lower salaries than tech although better than biologists, you will hit the ladder cap faster, very few transcend to become leaders and command multidisciplinary groups. Good luck!
- davidf18 10y ago
  You should work hard in your pre-med courses and get good grades, if you have not already taken them. The same is true for MCATs. I took Physical Chemistry which I am told is looked at by medical admissions offices and something which standard applicants would have a hard time with but those with math/engineering backgrounds should do well in.
- GarrisonPrime 10y ago
  Medicine was always my goal, so I decided I'd use my undergrad time as an opportunity to try something else I was always interested in and which could provide me with some radically different perspective compared to the biology/chemistry folks. (And would also serve as a backup if I failed to get into medical school...) My transition was easy. The required prerequisite college courses for admission to med school (two semesters biology, two physical chemistry, two organic chemistry, two physics, one basic calculus) were more than sufficient to get me through my med school courses. I took a few extras too (anatomy/physiology, molecular biology, and genetics) but I feel those were overkill and might not have been necessary. At least they were fun. I think the biggest part though was that I sought out some experiences in the medical field. This demonstrated to the admissions committees that my career goals weren't a lark and I knew what I was getting into. I did a 6-week summer mini-med-school camp, shadowed physicians in the ER a couple of times, and volunteered in a medical records department for a semester. When I didn't get into med school the first time, I spent the interim year getting my Certified Nursing Assistant license and working in an assisted living facility and a hospital's locked psychiatric ward. Fun times. Good luck.
evoloution 10y ago
You have a valid point. I felt that my love for math and algorithmic thinking (although not part of a formal curriculum) have been very helpful. Medicine is more like engineering than most people think. Keep in mind though that everything looks like a nail to a hammer. The same way they lacked in some way, you lack too. Unless you expand your knowledge base in the end you will be limited in analyzing similarly generated data with slightly better ways, as is so often seen in the literature. At the same time researchers who possess the knowledge to work "under the hood" are designing novel ways to obtain data, ask new questions and design the accompanying pipelines in the process. Where you put mathematics major feel free to insert molecular biologist/ nuclear physicist/ organic chemist/ or whatever else you like. I feel that if you really want to be spectacular and at the interface of many sciences you have to master one and have a good understanding of the others. And as a doctor you will always have it easier to lead a team for human disease research, the prestige of the degree by itself and the fact that you have actually seen what is being studied will always give you some respect if you are not a complete jerk. The bar can only be set higher and higher...
wavegeek 10y ago
> many of my classmates were essentially relying on memorization. This so much: Cookbook approach to statistics without insight leads to terrible results, much data effectively thrown away - very costly data $millions worth. I have some complex medical conditions and it is just terrible to see highly trained specialists who are nothing more than cookbook executors. Once you are off the beaten path they are at a complete loss. The rampant statistical illiteracy in the medical world from general practitioners to researchers make this era of big data a huge risk. <Person I know> is a molecular biologist and jokes that her medical practitioner friends know almost nothing about how the human body works. > Memorisation > Cookbooks > Linear regression, maybe
xaa 10y ago
Bioinformatician here. All my formal training is in biology, but I've spent most of my research hours learning CS, statistics, etc, rather than biology. > An astonishing amount of biology is rather mechanistic, systematic, and logical. I disagree vehemently. Or, to be more precise, it is very likely that this is true in reality, but in terms of the dominant way it's viewed in both clinical and research practice, it's very much not. The average paper's findings are qualitative, not quantitative, in nature. Of course, numbers, statistics, etc, will be used -- but they will be used to get p-values to support a qualitative conclusion, rather than to work towards a quantitative, systematic understanding of biology. There are exceptions, of course -- a decent amount of biochemistry is quantitative enough -- but when you get to topics like cell signalling, "cell biology", etc, it is almost never quantitative. I have a favorite question I like to use in seminars when someone is presenting findings along the lines of "protein A regulates protein B". The question is: "Among the various regulators of B, how important is A? Is A a key regulator of B, or simply one of many?". Or, even more simply, "What is the most important regulator of B?" I invariably get blank looks, as if this is an unreasonable or impossible question to answer. Of course it is, as long as your models are qualitative. The point of the question is to get the speaker to think about that, but so far, no luck.
- GarrisonPrime 10y ago
  I didn't meant to imply biological systems were simple A-->B boxes or machines with perfectly predictable actions and reactions. I worded my comment poorly. You are quite right that it's a dynamic, multi-faceted reality. I think the blank looks are saying that it's unreasonable to ask because it's impossible to know; all they're doing is trying to tease out at least one connection. Hopefully, none of them seriously think they've hit the nail on the head and have found the one true mechanistic answer.
  - xaa 10y ago
    > one true mechanistic answer Well yes, if they thought this, they would likely be wrong. But that isn't what I was suggesting. The answer I would be looking for would be something like "A accounts for approximately 20%/50%/90% of the variance of B in context C". Clearly many or most biological entities have more than one input and output. My problem isn't that there's complexity, it's that we don't/can't/don't even try to quantify and robustly model it. The reason why I feel it is such a problem is that without quantification of effect sizes, (and obviously these effect sizes would be probabilistic and perhaps context-dependent), we cannot really predict much of anything. Another way of making this point is to consider the sort of pathway or signal transduction diagram you often see in Nature Reviews, etc. Then imagine for a moment that you are looking at a electric circuit diagram instead of a biological diagram. You have a resistor here connected to that capacitor there, etc... What's missing? Resistances. Capacitances. More generally, numbers or functions specifying the behavior of the nodes and edges conditional on their inputs. Such an electric circuit diagram would be almost entirely useless to a person who wanted to predict the behavior of the circuit under different conditions. I guess I just object to the statement (and you are far from the only one to have implied it, if indeed that's part of what you were saying) that we have a mechanistic understanding of X, if we cannot predict the behavior of X conditional on perturbations, even approximately. For me, this was and is a major stumbling block in my ongoing journey to learn biology. I am often shown such diagrams and told "this is how it works", as if these diagrams had any explanatory or predictive power. I am perpetually confused about how and why people think they do. N.B. I admit they have SOME explanatory/predictive power, but it seems to me to be very close to zero. The only remotely quantitative aspects of these diagrams are the stimulatory/inhibitory "arrowheads", indicating a positive or negative correlation, and missing edges, indicating presumed conditional independence. These diagrams also lack any temporal information.
hilop 10y ago
Common wisdom is that for most STEM and professional graduate degrees, the best undergraduate preparation is a math degree and a minor in the related field. If you do a major in any science or social science, have the time is wasted on tiptoeing around the math the prof is afraid you can't handle. Exceptions: lab/field work aren't covered by math class (do these as your minor and senior year) , and theoretical physics/chem do the math -- but you are better off doing a year of math and a year of accelerated whatever, than doing intro whatever and then advanced whatever.
- GarrisonPrime 10y ago
  I agree with this assessment. I've given dozens of presentations to my fellow hospital docs, and by far the most attended and appreciated are those dealing with proper use and interpretation of statistics.

maxerickson 10y ago

But how did your biology undergrad go?

maxerickson 10y ago
Right, I forgot that sarcasm is the worst form of sarcasm. The GP is implicitly comparing the impact a math curriculum had on their reasoning with the impact that a biology curriculum might have had. But n=1 for the math curriculum (perhaps the beneficial traits are endogenous) and n=0 for the biology curriculum.

qud 10y ago

How did you transition from a Maths/Programming undergrad to doing medicine? I'm doing Computer engineering and plan to do go into medicine later, any tips?

evoloution 10y ago
If I were you: 1) I would go work in a computational biology lab, get some papers out, Cancer research is always a step ahead but there a lot of good projects in Neuroscience too (immunology could also work). Go to the best lab you have around, most PIs know that it is hard to find good programmers because they mostly go to the big tech companies or startups. Make sure you don't do wet lab at this step it is going to be a waste of your time since you will not have time to do anything meaningful. 2) With the papers apply to medicine or preferably to MD/PhD program, if you get into MD/PhD program you will have time to learn some real science too. In any case if you are the ambitious type of person when doing step 1 try not to get stuck working as a bioinformatician: lower salaries than tech although better than biologists, you will hit the ladder cap faster, very few transcend to become leaders and command multidisciplinary groups. Good luck!
davidf18 10y ago
You should work hard in your pre-med courses and get good grades, if you have not already taken them. The same is true for MCATs. I took Physical Chemistry which I am told is looked at by medical admissions offices and something which standard applicants would have a hard time with but those with math/engineering backgrounds should do well in.
GarrisonPrime 10y ago
Medicine was always my goal, so I decided I'd use my undergrad time as an opportunity to try something else I was always interested in and which could provide me with some radically different perspective compared to the biology/chemistry folks. (And would also serve as a backup if I failed to get into medical school...) My transition was easy. The required prerequisite college courses for admission to med school (two semesters biology, two physical chemistry, two organic chemistry, two physics, one basic calculus) were more than sufficient to get me through my med school courses. I took a few extras too (anatomy/physiology, molecular biology, and genetics) but I feel those were overkill and might not have been necessary. At least they were fun. I think the biggest part though was that I sought out some experiences in the medical field. This demonstrated to the admissions committees that my career goals weren't a lark and I knew what I was getting into. I did a 6-week summer mini-med-school camp, shadowed physicians in the ER a couple of times, and volunteered in a medical records department for a semester. When I didn't get into med school the first time, I spent the interim year getting my Certified Nursing Assistant license and working in an assisted living facility and a hospital's locked psychiatric ward. Fun times. Good luck.

evoloution 10y ago

You have a valid point. I felt that my love for math and algorithmic thinking (although not part of a formal curriculum) have been very helpful. Medicine is more like engineering than most people think. Keep in mind though that everything looks like a nail to a hammer. The same way they lacked in some way, you lack too. Unless you expand your knowledge base in the end you will be limited in analyzing similarly generated data with slightly better ways, as is so often seen in the literature. At the same time researchers who possess the knowledge to work "under the hood" are designing novel ways to obtain data, ask new questions and design the accompanying pipelines in the process. Where you put mathematics major feel free to insert molecular biologist/ nuclear physicist/ organic chemist/ or whatever else you like. I feel that if you really want to be spectacular and at the interface of many sciences you have to master one and have a good understanding of the others. And as a doctor you will always have it easier to lead a team for human disease research, the prestige of the degree by itself and the fact that you have actually seen what is being studied will always give you some respect if you are not a complete jerk. The bar can only be set higher and higher...

wavegeek 10y ago

> many of my classmates were essentially relying on memorization. This so much: Cookbook approach to statistics without insight leads to terrible results, much data effectively thrown away - very costly data $millions worth. I have some complex medical conditions and it is just terrible to see highly trained specialists who are nothing more than cookbook executors. Once you are off the beaten path they are at a complete loss. The rampant statistical illiteracy in the medical world from general practitioners to researchers make this era of big data a huge risk. <Person I know> is a molecular biologist and jokes that her medical practitioner friends know almost nothing about how the human body works. > Memorisation > Cookbooks > Linear regression, maybe

xaa 10y ago

Bioinformatician here. All my formal training is in biology, but I've spent most of my research hours learning CS, statistics, etc, rather than biology. > An astonishing amount of biology is rather mechanistic, systematic, and logical. I disagree vehemently. Or, to be more precise, it is very likely that this is true in reality, but in terms of the dominant way it's viewed in both clinical and research practice, it's very much not. The average paper's findings are qualitative, not quantitative, in nature. Of course, numbers, statistics, etc, will be used -- but they will be used to get p-values to support a qualitative conclusion, rather than to work towards a quantitative, systematic understanding of biology. There are exceptions, of course -- a decent amount of biochemistry is quantitative enough -- but when you get to topics like cell signalling, "cell biology", etc, it is almost never quantitative. I have a favorite question I like to use in seminars when someone is presenting findings along the lines of "protein A regulates protein B". The question is: "Among the various regulators of B, how important is A? Is A a key regulator of B, or simply one of many?". Or, even more simply, "What is the most important regulator of B?" I invariably get blank looks, as if this is an unreasonable or impossible question to answer. Of course it is, as long as your models are qualitative. The point of the question is to get the speaker to think about that, but so far, no luck.

GarrisonPrime 10y ago
I didn't meant to imply biological systems were simple A-->B boxes or machines with perfectly predictable actions and reactions. I worded my comment poorly. You are quite right that it's a dynamic, multi-faceted reality. I think the blank looks are saying that it's unreasonable to ask because it's impossible to know; all they're doing is trying to tease out at least one connection. Hopefully, none of them seriously think they've hit the nail on the head and have found the one true mechanistic answer.
- xaa 10y ago
  > one true mechanistic answer Well yes, if they thought this, they would likely be wrong. But that isn't what I was suggesting. The answer I would be looking for would be something like "A accounts for approximately 20%/50%/90% of the variance of B in context C". Clearly many or most biological entities have more than one input and output. My problem isn't that there's complexity, it's that we don't/can't/don't even try to quantify and robustly model it. The reason why I feel it is such a problem is that without quantification of effect sizes, (and obviously these effect sizes would be probabilistic and perhaps context-dependent), we cannot really predict much of anything. Another way of making this point is to consider the sort of pathway or signal transduction diagram you often see in Nature Reviews, etc. Then imagine for a moment that you are looking at a electric circuit diagram instead of a biological diagram. You have a resistor here connected to that capacitor there, etc... What's missing? Resistances. Capacitances. More generally, numbers or functions specifying the behavior of the nodes and edges conditional on their inputs. Such an electric circuit diagram would be almost entirely useless to a person who wanted to predict the behavior of the circuit under different conditions. I guess I just object to the statement (and you are far from the only one to have implied it, if indeed that's part of what you were saying) that we have a mechanistic understanding of X, if we cannot predict the behavior of X conditional on perturbations, even approximately. For me, this was and is a major stumbling block in my ongoing journey to learn biology. I am often shown such diagrams and told "this is how it works", as if these diagrams had any explanatory or predictive power. I am perpetually confused about how and why people think they do. N.B. I admit they have SOME explanatory/predictive power, but it seems to me to be very close to zero. The only remotely quantitative aspects of these diagrams are the stimulatory/inhibitory "arrowheads", indicating a positive or negative correlation, and missing edges, indicating presumed conditional independence. These diagrams also lack any temporal information.

hilop 10y ago

Common wisdom is that for most STEM and professional graduate degrees, the best undergraduate preparation is a math degree and a minor in the related field. If you do a major in any science or social science, have the time is wasted on tiptoeing around the math the prof is afraid you can't handle. Exceptions: lab/field work aren't covered by math class (do these as your minor and senior year) , and theoretical physics/chem do the math -- but you are better off doing a year of math and a year of accelerated whatever, than doing intro whatever and then advanced whatever.

GarrisonPrime 10y ago
I agree with this assessment. I've given dozens of presentations to my fellow hospital docs, and by far the most attended and appreciated are those dealing with proper use and interpretation of statistics.