More Recent Comments

Monday, March 09, 2015

Learn to think like a scientist

There's a course at MIT (Boston, USA) called "7.00x Introduction to Biology - The Secret of Life." It's a very popular MOOC (online course). Here's the trailer for the course. In it, Eric Lander tells you that if you take his introductory biology course you will learn to think like a scientist and you will be able to understand the latest breakthroughs.


Here are the week one lectures that focus on biochemistry. I don't have time to go through it all but check out the description of ATP beginning at 2:13. This is not how good teachers explain the importance of ATP in the 21st century but it is how it was taught 40 years ago.


Here's the week two lectures (below). Check out the part at 2:08 where Eric Lander is talking about the "Energetics of Pathways." See if you agree with his explanation and his references to entropy. Would you explain this without talking about forward and reverse rates?

The next section on "Tricks of Pathways" is very interesting. Keep in mind that he is discussing "tricks" that make glycolysis work but most of those reactions also work in reverse to make glucose (gluconeogenesis). This gets to be a problem with trick #2 at 2:25 when he says that an "unfavorable" reaction can be "pulled" in the right direction if the next reaction in the pathway is very favorable. This explanation was popular many decades ago but now we know that almost all the reactions are at equilibrium and ΔG = 0.

My point is that just because MIT is a prestigious university and Eric Lander is a famous scientist, does not mean that this is the best undergraduate course and the best way to teach biochemistry. Some of my colleagues at colleges you've never heard of could do a much better job. Eric Lander could do a better job if he would just read a modern textbook.




26 comments :

Robert Byers said...

If one can learn the secret of life, kids, then its not a secret!
Think like a scientist! Here we go again.
All a scientist is IS a person applying a superior methodology before conclusions are drawn.
Otherwise there is no difference in thinking.
In the video it seems its hit about cool things in the news.
Yet why can't a wiki article do the same thing.?
i wish them well but its odd to see the advertising stresses.
Did the Genome project ever heal anyone.?i hope so but mapping it seems a minor thing .

frankly said...

He posted Objectives. That pedagogy is 40 years old, too.

ProfofBC said...

Breaking bonds releasing energy is just a horrible way to teach students and confuse them later on. Not to mention focusing only on the negative charge, ignoring resonance, salvation effects, and concentrations, doesn't set-up students well for the rest of metabolism.

The entropy discussion was just bad. I don't think the professor understands standard Gibbs energies (biological or otherwise) or at least didn't convey that understanding. Also label your axes for the G3P DHAP example. Kelvin not degrees Kelvin. Minor relative to the bigger issue of getting the students to understand delta Gibbs energies, equilibrium, and relation to entropy.

How in the world can reaction A to B be coupled to reaction C to D? That won't accomplish anything that is being suggested.

Explaining "trick 2" properly means getting the students to understand the underlying thermodynamics which obviously from before didn't happen.

The glycolysis is just bad and setting up misconceptions that the biochemistry professor later on is going to have correct and that is extremely challenging. Energy out of the bonds is such a bad way to have students understand what is going on. And no, the chemical structures are important for glycolysis.

What a boring class. Where is the constructivist approach having students build upon previous knowledge/understanding?

Larry Moran said...

You and I are among the few people who recognize that this is horrible teaching. Unfortunately, the myth persists that because this is an MIT course it must be among the best in the world.

What surprises me more than anything is the students in these classes. It looks like they've never heard of amino acids or glycolysis before taking this class. It looks like the thermodynamics they must be getting in their chemistry class doesn't transfer to this class.

These are supposed to be among the brightest students in the United States but not one of them challenges even the most obvious errors by the lecturer. You see the same thing when you watch lectures at Harvard on video. What the heck is going on?

Larry Moran said...

How in the world can reaction A to B be coupled to reaction C to D? That won't accomplish anything that is being suggested.

This is standard 1960s biochemistry. It conveys the impression that an enzyme catalyzes the hydrolysis of ATP, for example (C to D) and then "captures" the energy released and uses it to force A to B. That never happens. Enzymes can't do that.

These are the sorts of concepts that have to be taught correctly the first time because it will be much more difficult to correct a misconception later on. Fortunately (?), the MIT students won't have that problem because none of their other lecturers will ever talk about it again and, if they do, they probably don't know the correct concept anyway.

Remember, these students are the future leaders of ENCODE projects. It explains a lot.

judmarc said...

These are supposed to be among the brightest students in the United States but not one of them challenges even the most obvious errors by the lecturer. You see the same thing when you watch lectures at Harvard on video. What the heck is going on?

These students are practicing skills they will need for corporate life. No one gets ahead by challenging those in charge.

ProfofBC said...

It is an intro bio course, the first they take at the university level. There are no pre-reqs. MIT students likely did take AP Biology and AP Chemistry in high school but these courses are typically taught to the test and students study to get a high score not to learn. This approach means they compartmentalize what they do learn. The teaching style also doesn't encourage students to challenge and reinforces the compartmentalization approach to their learning.

Ideally university chemistry would be a pre-req for the intro to cellular biochemistry or have an integrated course where the chemistry is taught through biological examples by experts who can actually guide the students to learn and how to leverage understanding to new situations.

Where I currently teach, intro cellular biochemistry also is taught in the biology department and doesn't have a pre-req. Similar poor teaching occurs and it is a problem when they take biochemistry later on. In some ways those who haven't had biology are actually in a better position for the biochemistry course as they haven't been taught wrong.

Larry Moran said...

Our students have to take two biology courses in their first year. One semester is entirely devoted to evolution and the other is cellular & molecular biology. They take biochemistry in second year and the prerequisites include a semester of organic chemistry and a semester of physical chemistry that must be taken in first (freshman) year.

The second year biochemistry course is a prerequisite for almost all upper level courses.

They will have learned about the fundamentals of biochemistry in high school including basic metabolism and the structures of biomolecules. That isn't always taught correctly so we have to concentrate on fixing the misconceptions.

I must admit, reluctantly, that there are lecturers in my own department who are no better than Eric Lander. I would like to fix that.

roger shrubber said...

I was waiting for the obvious response. If so many can do better, why haven't they?
Are you especially opposed to the medium? I have problems with the idea of learning by watching a video. People who believe this is an effective alternative to a proper education are deluded, and yet, that perception exists, and it will grow.
Which is the bigger problem? The abused biochemistry or the medium?

Anthony Chiu said...

You've forgotten that every student at MIT has to take it, as does Harvard. So you will have people who have never heard of glycolysis or amino acids, because that is a grade 12 level concept. You also have students who do not care at all how it works; they just want to pass. Eric and Bob have to explain the concepts in a way that is understandable to people who don't know the basics while hopefully be engaging, so you will see that he simplifies some of his descriptions of biological processes. The students you see are different at UT. BIO150 is taken primarily by premeds and future grad students in Biology; not by commerce kids and English majors. In contrast, here at MIT, intro bio is one of 4 courses that must be taken to graduate.

Georgi Marinov said...

That is true, however, that does not mean the MIT Biology programs is flawless. I don't know how exactly it is at the moment, but a few years ago when I passed through it, there were many terrific courses, but it was also true that some of the basic courses were really rushed and everyone would have benefited from a slower and more in depth treatment of the subject. This applies for the general genetics course (7.03) which was just one semester and did not cover anywhere nearly as much as every biologist should know about the subject, and it was especially true about the biochemistry class (7.05), which when I took it, consisted of one semester half of which was spent on DNA and protein structure and the other half on metabolism. You cannot learn (the basic facts), let alone fully understand metabolism in half a semester. That should be spread over a whole year and I can only tell that because that was not my first exposure to university-level biochemistry (which is not the case for most people). Now, at least when it comes to biochemistry, there was a more advanced course to fill in the gaps, but there was also another, this time truly enormous, gap in the program, and it was the complete absence of any course on evolution. Which is simply unbelievable for an institution of that magnitude when you think about it. Sure, you have a bunch of world class evolutionary biologists a 20-minute walk to the north and you can take classes there, but it's not as if anyone was ever informed that they need that knowledge and encouraged to do that.

Now why is there no course on evolution? Because there is no evolutionary biologist on the faculty to teach it. And why that is should be obvious to anyone who understands how the system works..

Which is part of the fundamental point the original post was getting at - at the moment world-class cutting edge research and correspondingly good teaching have increasingly gone their separate ways. That's not necessarily the fault of any individual, it's just how the incentives are stacked. There are only 24 hours in a day. The question is was that always the case. The math and (theoretical) physics departments may be the closest thing to a control group one could find at present - mathematicians have been under a lot of the same institutional pressures as biologists have, but at least they don't have 25-person labs to supervise so one would think that if there ever was a golden time when the best research and the best possible teaching went hand in hand, its vestiges would be found there.

P.S. I have in fact always been amazed at the fact that Lander and Weinberg are teaching that course - most people in their position can easily get away from their teaching obligations by organizing a seminar class where they discuss current research (and that would be an amazing seminar for the people who have the background to take advantage of it - I've been in a few of those and that has been my experience) and don't have to think about how to explain stuff from which they have moved on decades ago to first-year undergrads (which is not at all an easy task). But it's great that they do that because their presence alone helps attract many students to the subject.

Georgi Marinov said...

Laurence A. MoranTuesday, March 10, 2015 3:27:00 PM

These are the sorts of concepts that have to be taught correctly the first time because it will be much more difficult to correct a misconception later on. Fortunately (?), the MIT students won't have that problem because none of their other lecturers will ever talk about it again and, if they do, they probably don't know the correct concept anyway.


That is in fact correct - they won't. Most more advanced classes have no relation to these things.

The unfortunate fact is that a lot of biology can be learned and done without any knowledge of biochemistry. It doesn't mean it should be that way, but that's how things are.

Remember, these students are the future leaders of ENCODE projects. It explains a lot.

You are unfairly singling out ENCODE. It is true that many people in genomics know nothing about biochemistry (and not just genomics - there is a long list of fields in biology where one can do good research without knowing much about it), but this is not because they have been taught biochemistry badly, they have simply never been taught any biochemistry because they come from completely different backgrounds - computer science, engineering. physics, math. Again, that is not ideal, but is still an unfortunate fact of life - because biology departments have never considered it necessary to seriously develop any substantial computational skills in their students, at the research level, that expertise has to be imported from somewhere else. And this situation will not improve significantly in the foreseeable future, because the incentives are once again not there - a token bioinformatics course here and there is not the same as real computational training. But real computational training takes time and effort on the part of the faculty, and time and effort on the part of the students. Which nobody can afford to expend - grants and papers have to be written (by the faculty) and experiments have to be done at the bench (by the students) and there isn't time and energy for much else.

There is, of course, a lot of time for teaching math and statistics in K12, when the workload is generally very light (stupid tests aside), but that's way too heretical of an idea...

Georgi Marinov said...

P.S. When I say "without any knowledge of biochemistry" I don't mean that literally, I mean deeper understanding.

ProfofBC said...

Eric and Bob have to explain the concepts in a way that is understandable to people who don't know the basics while hopefully be engaging, so you will see that he simplifies some of his descriptions of biological processes.
***********************************************
Anthony being wrong isn't a simplification. What we are talking about is explaining the material incorrectly reinforcing misconceptions not correcting misconceptions. That is not good teaching. In addition, the approach isn't an engaging one nor one that empowers students.

Larry Moran said...

@Georgi

Your comments are very informative and helpful. Thank-you very much.

I don't think it was unfair of me to single out ENCODE leaders as a prime example of research scientists who lack in depth understanding of topics like biochemistry and evolution. As you point out, many of those leaders are more interested in the technology than the theory and many of them have a background in computer science and not biology.

I understand that. They should recognize their limitations and not make sweeping pronouncements about things outside of.their area of expertise. They are behaving just like Eric Lander. They think they understand biochemistry based on things they may remember from many years ago. In Lander's case he thinks he understands it well enough to teach a course. In the case of ENCODE leaders they think they understand it well enough to publish their conclusions in Nature.

ProfofBC said...

Larry how well does the intro cellular & molecular biology course do?

Our students take a semester of intro physical chemistry and then two of organic chemistry before biochemistry so it is a spring semester second year course here. The biology department and other life science departments would love biochem with only a one semester organic requirement (they would prefer none) but our students typically have 0 organic background from high school. Those that have had organic in high school, typically don't retain much of it.

The first semester biochemistry course is a pre-req for all other biochem courses but the cellular/molecular courses in biology don't require it. Most of the biology students really enjoy biochemistry because it gets them to understand how and why the processes they discuss in their cell/molecular biology courses occur.

Larry Moran said...

@Anthony Chiu

I did not know that ALL students have to take this course. That doesn't excuse the lecturers because once they've decided that commerce students and English majors need to understand glycolysis and thermodynamics they still need to teach it correctly.

It does, however, raise some other concerns. If you have a captive audience consisting of every single student at MIT then why are you teaching that material in the first place? You should be concentrating on evolution.

And why in the world would the lecturer make the outlandish claim that hiis English majors will learn to think like scientists and be able to understand the latest research when they finish the course?

We can't even achieve those objectives with students who MAJOR in biochemistry!

Georgi Marinov said...

Laurence A. MoranWednesday, March 11, 2015 7:18:00 AM
@Anthony Chiu

And why in the world would the lecturer make the outlandish claim that hiis English majors will learn to think like scientists and be able to understand the latest research when they finish the course?


That is indeed not going to happen based on this course, but something that the MIT biology program is in fact quite strong in is the "thinking like a scientist" part. It varies from course to course, but from what I have seen of curricula elsewhere, it is very rare to put so much emphasis on that. There is a bit of it in the intro class too, but it is the advanced classes that really emphasize it. Exams mostly consist of putting you in some experimental situation and you having to figure out what's going on (and it's typically not trivial). In the best classes I took it wasn't even hypothetical situations - it was real data from papers and you had to derive conclusions from it.

Now, this can be and sometimes is taken too far, at the expense of actually teaching core concepts - sometimes it becomes an exercise in "We''ll teach you how to figure out things on your own but we won't teach you anything concrete that will last".

So there is room for improving the balance, but the fact that the thinking-like-a-scientist component is very heavily emphasized remains

Georgi Marinov said...

Laurence A. MoranWednesday, March 11, 2015 7:01:00 AM
@Georgi

As you point out, many of those leaders are more interested in the technology than the theory and many of them have a background in computer science and not biology.


One more thing - methods and technology matter. A lot.

So someone has to be interested in those - all that talk about "mere technicians", etc. is actually just as poisonous as the excesses of focusing on technical wizardry at the expense of deep ideas because it is very easy to misunderstand it (and I am not even sure it's always a misunderstanding) as saying that the technical details are irrelevant. For every paper that messed up its conclusions because of poor understanding of the theoretical and historical background there is another that is ruined by poor understanding and/or application of the methods it is based on leading to improper biological interpretations.

Both things need to paid sufficient attention.

Larry Moran said...

@Georgi

I'm not disagreeing with you about the importance of technology. However, technical skills are the SERVANT of science, not it's master. It doesn't matter how skilful you are at finding transcription factor binding sites if you don't know why you are doing the experiment in the first place and how to interpret the results. Theory and hypotheses drive the scientific quest for knowledge, not just the ability to do something well.

ENCODE scientists spent a lot of time and money developing technical skills and complex algorithms to map all binding sites for dozens of transcription factors. Meanwhile, they seem to have been unaware of the fact that most of those don't mean anything and that spurious binding is exactly what you expect of DNA binding proteins and a genome full of junk. It's a question of priorities and they blew it by concentrating too much of the trees while losing sight of the forest.

Larry Moran said...

@Georgi,

Learning to think like a scientist means understanding the nature of science and the importance of evidence-based reasoning. It means learning how to distinguish good science from psedoscience and it means learniing how to think critically and be appropriately skeptical.

You don't learn these skills very effectively by reading the latest papers in the scientific literature and gaining an appreciation of the latest technologies and their limitations. You learn them by discussing controversies and misconceptions. The best thing MIT could do in 2015 is teach a class on genomes and junk and let the students discuss why the ENCODE scientists were wrong.

I'm pretty sure that's not going to happen unless you get a job there.

Anthony Chiu said...

@Larry Moran

First, there are 3 bio courses students can take. 7.012 is this one, and involves the most cell biology. 7.013 is focused more on human disease whereas 7.014 involves more evolution and ecology. Knowing that most students are not going to study bio beyond this point, the instructors have a responsibility to make sure the students are able to understand what a stem cell is, what is cancer, etc so when they read newspaper articles, they can properly assess it. Some evolution is covered in all 3 (1-2 classes I think), so that students understand where antibiotic resistance is coming from. However as instructors, we have a responsibility not to teach everything possible but what is most relevant in the current day. There is no need to teach Hardy-Weinberg in an intro class unless they plan on going into ecology because otherwise, they will forget it.

Secondly, we do teach students the beginnings of thinking like a scientist. The very first class focuses on the distinction between how geneticists and biochemists think (removing things vs trying to reconstitute things). The exam problems we write are literally interpret this restriction digest gel, what does this RNA-seq data tell you based on what you already know about Kras (for example)? I personally think testing this way is a much better way to teach biology than say regurgitation heavy classes because the ability of applying ones knowledge to novel situations is shows actual learning. This is actually what MIT does for every class starting first year with their biweekly problem sets.

Georgi Marinov said...

Laurence A. MoranWednesday, March 11, 2015 9:40:00 AM
@Georgi,

Learning to think like a scientist means understanding the nature of science and the importance of evidence-based reasoning. It means learning how to distinguish good science from psedoscience and it means learniing how to think critically and be appropriately skeptical.


It is closer to your ideal than you think. It's just that you can only evaluate it well by spending some time with the courses (and online may not be the optimal medium for that)

Also, keep in mind that MIT has a requirement that in order to graduate you need to take 8 humanities classes. Which exactly, you have some flexibility in choosing, and many people don't choose as wisely as they should, but the options include some very useful stuff such as scientific writing at a very good level, and history and philosophy of science. Which is quite heavy on historical and current controversies, epistemology, etc. I absolutely hated writing the papers for those classes but loved being in them and learned a lot. That should probably be institute-wide requirement, but that is in an ideal world that does not and cannot exist in practice - in reality these courses only work in relatively small groups with the instructor, and there is no way to provide that for everyone when the size of each class is 1200.

Jonathan Badger said...

Of course thinking like a *modern* scientist means understanding computational analysis and quantitative reasoning, which is something a lot of experimental biologists don't have in my opinion. And in this era of big data, they need to.

While I don't agree with ENCODE's definition of "function" either, I think the whole argument against the leaders that they are "just computer scientists" is wrong-headed. It's rather reminiscent of how when I was in grad school in the 1990s there were elderly ecologists and such that sniffed at molecular biologists and said they were mere gene-jockey technicians. Well, those people eventually died off and the new generation of ecologists understand that molecular biology is crucial for their science. The same will happen for bioinformatics.

Robert Byers said...

Again . What is the nature of science? What is the nature of a vern?
Science can only be if its about very careful investigation, into something, and abnd only after this making conclusion.
Science is presented to the public as a higher standard of investigation and so it can demand confidence in its conclusions. other subjects demand for confidence is less legitimate but still pretty good.
Science must be a vigorous verb for investigating nature or its nothing in nature any different then anything done by man with skill.
So yes studying psudoscience is useful but also studying if evolution is a scientific theory is useful too.
In fact I don't see that science has much to do with these intro subjects for kids.
They are just learning, largely memorizing, the conclusions from science researchers in the past.
these kids just need to know results and thats all they are intellectually able to do.
Actual scientific thinking/research is done by very few people.
95% will never think like a scientist. Most scientists don't as this creationist bumps into.
At best science is a methodology that firects human thinking.

Anthony Chiu said...

@Jonathan
I completely agree. As of now, quantitative reasoning and computational analysis is just as essential in modern biology, as is understanding the concepts of genetics and biochemistry. Mainly because genomic/proteomics studies allow for unbiased discovery, so you can pick out new hypotheses that you would not have known using prior knowledge alone (which was the old way of doing biology).

However, the problem with modern day biology is many biologists are trying to analyze computational data without understanding what is going on. As a researcher, you wouldn't want to interpret data without deeply understanding the technique. For instance, someone who can run a Western blot but doesn't know the difference between native vs denaturing gels will interpret things differently. Likewise, how you normalize your RNA-Seq data leads to drastically different interpretations: every gene could be changing or none. While most biologists can intuitively understand wet lab techniques, dry lab techniques are not as easily understood, which is why all programs really need to encourage quantitave understanding early on so that if they do end up doing those techniqques, they won't be lost. If I could repeat college all over at UT, I would have dropped some BCH classes to focus on learning computational/quantitative stuff because learning it now in grad school is putting me at a competitive disadvantage.