Pipes and Moving Staircases

One of the challenges of embarking on a novel research project is trying to clarify what exactly I’m trying to accomplish as I delve into the details. I know the big picture: My goal is to understand how proto-metabolic systems self-assemble from chemical mixtures of simpler organic molecules. The approach I’m exploring is to first learn some new methodology, but not use it in the way it was originally developed. Instead, I plan to combine it in a novel way with familiar-to-me methodology, and hopefully (fingers crossed) come up with some models to test.

That’s still very broad. Too broad. My collaborator challenged me to try and articulate more clearly what I thought the model needed, how I planned to use it, and what I hoped it would be able to predict more concretely.

I’d been mulling it over for several days while haphazardly typing down seemingly random thoughts in the hope that some catalytic mechanism would crystallize into a model. Ha! That sounds like the actual problem I’m trying to solve chemically.

Then two images came to my mind.

The first was the word “PIPES”. In fact, I was picturing a scrawled piece of paper dug out from the hand of Hermione when she had been petrified in Harry Potter and the Chamber of Secrets. Not quite the image below, which I think is from the movie prop; I couldn’t find anything on the Internet that came close to matching the image in my mind.

That led me to envision a complex network of pipes with many intersections and meeting points, converging and diverging, going up and down, snaking left and right. A spigot began the flow of water at the input, and the water flowed into many streams – the flux of water molecules changing over time as the complex network was traversed. Some pipe locations eventually ran dry while other flows grew into larger torrents, even widening the pipes as molecules rushed by. At the crux of metabolism is the throughput of the molecules of life – flowing through chains and cycles of biochemical transformations.

But this picture was missing a crucial piece needed to allow such systems to evolve.

A day later I was catching up with friends and I was telling them how I got lost in a part of campus I was unfamiliar with. One of them remarked that she had also found it confusing and was ‘convinced’ that the staircases moved, a la Hogwarts. (We’d had a number of amusing Harry Potter related discussions in years past!)

One more day passed before it hit me that some of the pipes in my network should be allowed to move over time as the system evolves. Thus, the second image was ‘moving-staircase’ pipes. New connections made and old connections broken! As new molecules are made, some of which can catalyze prior chemical reactions, they can alter the flows by providing alternate pathways.

I now have a model in mind, though I need to flesh out more details. The model may prove less than ideal as I flesh things out – all models are wrong, some are useful – but it’s a good start. Who would have thought that pipes and moving staircases from the Harry Potter books would come in handy; you might say it was magical!

Prime Matter: Pantogen

My introductory chemistry courses begin with the following question: “What is matter and why does it matter?” We consider several ideas from the ancient Greek philosophers, before a final compare-and-contrast of Democritus’ Atomic theory and the Four Element system championed by Aristotle. Is matter fundamentally made up of atoms – discrete indivisible particles too small to observe? Or is matter a continuous blend of principles – embodied by the ‘elements’ earth, water, air, fire? In the alchemists' view, matter can be transmuted – you could turn lead into gold with the right ingredients in the right proportions.

We then jump almost two thousand years to Dalton’s Atomic theory, and define terms that are used in modern chemistry: element, atom, molecule, compound. The periodic table is introduced and elements are distinguished by their number of protons. Since the protons reside in a tiny nucleus not subject to chemical reactions (which involve only the movement of electrons on the atoms periphery), transmutation is very difficult unless very high energies are involved that allow nuclear reactions to take place. We do discuss spontaneous transmutation via radioactive decay.

Once the basics of how electrons arrange themselves in atoms is established, we can start discussing chemical properties of elements and how they are arranged in the periodic table. The history of how the periodic table came to be has many twists and turns. Among several early attempts at formulating periodic law that I discuss briefly in class, one that throws the students for a loop (pun intended) is Hinrichs’ spiral. However, I had never delved further into his other ideas until reading about Hinrichs in The Lost Elements.

Apparently in the mid-1850s, Hinrichs developed a theory of prime matter based on a universal element he called pantogen. According to Hinrichs, what we call chemical elements are simply different combinations of pantogen. In his scheme, pantogen had a relative atomic weight of 1/128 compared to hydrogen. In a nod to Lavoisier, he utilized precision measurements of experimental relative atomic weights to make his case: “…if, from a liter of pantogen weighing 0.697 mg, one were to subtract the observed experimental weights of 1 liter of O, H, N, and C (gases), the new atomic weights of these elements would turn out to be the whole numbers 16, 1, 14, and 12, respectively.” (These are the atomic numbers of those four elements.)

One thing I try and emphasize in introductory chemistry is the strangeness of our model of the atom. Hydrogen, the simplest atom, has a positively charged proton that’s tiny in size, yet 1840 times more massive than the negatively charged electron that roams around in mostly empty space – like a bee in a cathedral. That’s actually a very counter-intuitive idea. Hinrichs thought that hydrogen is made up of 128 atoms of pantogen. He didn’t know about protons and electrons. However, his idea isn’t too far-fetched given that after protons and electrons were discovered, there was a proposal that the hydrogen nucleus was actually made up of approximately 1840 ‘plus-trons’ and 1839 electrons; a single electron happened to have escaped from the nucleus. (I named these hypothetical particles ‘plus-trons’ so as not to confuse them with positrons.)

The pantogen scheme did not catch on, apparently because very few people actually read Hinrichs’ work. You, dear reader, have probably never heard of pantogen. Apparently very few people read the work of Nikolai Morozov either, because otherwise our subatomic elements might be anodium, cathodium, and archonium – instead of the proton, electron, and neutron. A Google search on any of these terms gives you results unrelated to the theory of matter. I suppose they no longer matter.

Bumbling in Research

Learning some completely-new-to-me research methodology has been an interesting experience thus far. I would say that I’m bumbling along with the attendant feeling of being unsure of what I’m doing. I’d like to think that in computational research, bumbling around is part of the learning experience. It’s certainly the feeling I had when I first started undergraduate research. I had the same feeling when I started my first graduate school research project, and then again at the beginning of my postdoc. It didn’t bother me then, possibly because I was young, idealistic and quite willing to challenge myself to learn some new things. It’s been a while since I’ve had that experience, but I’m having it now.

While it’s frustrating to feel like a bumbling fumbler, I’m thankful for the experience because it reminds me how undergraduates feel when joining my research lab for the first time. Students almost always have no background in Linux or using command line text-editors, and certainly have not learned how to use the computational chemistry programs that are the bread-and-butter of my research projects. My students begin with sequenced tutorials and I slowly walk them through the nuts and bolts of how to start running calculations. I tell them that they will feel like bumbling fumblers, and that this is a completely normal experience. Midway through the semester they will realize they have become more proficient, and the number of errors they make will (hopefully) reduce steadily. Of course, telling them this means that while I intellectually understand what they will go through, I don’t actually experience the same feelings of confusion.

Except now I’m going through that bumbling experience, and it’s a good thing, even if it doesn’t feel as good right now. Thankfully I have a sabbatical to fumble my way through. This allows me to be in a relaxed frame of mind so the bumbling experience is much less stressful. I can take frequent breaks whenever I hit a wall, and work on something else that helps me feel more productive. When I feel ready, I can switch back to learning and bumbling along. Since I will not be tested on what I’m learning and there are no deadlines, you could say that my motivation for learning is completely intrinsic. (As a full professor, I don’t feel external pressure to be highly research productive; although my average publication rate remains unchanged since I was promoted.)

Another thing that helps me stay motivated is that learning this new-to-me research methodology might lead to a new approach to tackling a thorny and unsolved question in origin-of-life research: How did metabolism arise? Essentially, I’m focusing on the chemistry of proto-metabolic systems. Not much is known about this area and my hope is to marry the new-to-me methodology with old-hat-to-me methodology in a novel/creative way to tackle the problem. Much of my career has been spent working on projects that I’m fairly sure are going to be successful. (Not everything works, but much does.) That’s a safe way to keep publishing scientific articles, something that gets scrutinized when promotion-time comes around. Low-risk projects are also very suitable for undergraduates who have no research experience in my field; they help build student confidence and there’s less bumbling. On the other hand, more ‘open’ blue-sky or high-risk high-payoff projects tend to be less suitable for undergraduate research.

Maybe I’m telling myself all of this to stay motivated when I feel confused and less productive. It seems to be working: the motivation, not the project I’m working on – at least, not yet. Or maybe I’m building self-efficacy rather than motivation. In any case, I keep bumbling along.

P.S. In the meantime, I’m beefing up my new language learning while on sabbatical far from home, and it’s been easier this second time around!

Ultrasonic Brain Control

The Conversation had an article this week titled “Remote control for brain cells: scientists use ultrasound waves to activate neurons”. It introduced me to the field of sonogenetics, somewhat related to the more widely-known optogenetics except sound is used as the stimuli instead of light. Apparently, there is a protein present in some worm neurons that responds to pressure waves in the ultrasonic range. There’s even a video showing how “ultrasound pulses could make the worms change direction, as if we were using a worm remote control."

According to the article, human neurons do not have the ultrasound-sensitive protein, however it can be introduced via an engineered virus that “delivers genetic material to brain cells.” One advantage of using sonogenetics is its potential ability to target brain cells in specific areas; and this could be very helpful in dealing with movement disorder related brain diseases. If you wanted to get dystopian, you might imagine an evil scientist controlling your physical movements via ultrasound affecting your brain cells! It’s mind control of a sort.

This reminds me of the Imperius Curse in the Harry Potter books. While I have previously speculated on the use of electromagnetic radiation as a vehicle for magic, I suppose this can be extended to ultrasound. Both have wave-like properties, but ultrasonic waves can be transmitted through or into the body more easily than optical waves. However, if used as a scanner (the most common medical use you might be familiar with), there is a tradeoff between how far the waves can penetrate the human body and how good you need the resolution of the scan to be. Presumably if you’re manipulating neurons, you need fine control, so you’d better be close by. That could be a reason why the closer you are to the ‘victim’, the more effective the Imperius curve. Sound wave amplitudes also die with distance, in any case. Another drawback if ultrasound is the carrier wave, is that the Imperius Curse won’t work well in outer space. I suppose a wizard in space could resort to some other curse that is carried by electromagnetic radiation!

In Harry Potter and the Half-Blood Prince, an attempted Imperius curse goes awry, and the victim, junior minister Herbert Chorley, has his brain unfortunately addled – he thinks he’s a duck and he is committed to St. Mungo’s. This perhaps illustrates the challenge of manipulating neurons via ultrasound, certainly true in our science-that-sounds-like-magic world. You might say the same of fMRI, perhaps the equivalent of the Legilimens spell.

Science at a Liberal Arts College

I’ve had a number of conversations the past several weeks about how science departments operate in Selective Liberal Arts Colleges (SLACs) versus Research-Intensive Institutions (R1s). I will limit discussion to institutions in the U.S. since that’s the context I’m most familiar with.

What is a SLAC? The ‘Oberlin Group’ will give you a sense of the institutions that fall into this category. At present, they have an average size of 2000 students, and most range between 1600 and 2400. They are private institutions, expensive to attend, although often generous with financial aid. They typically do not have graduate programs (if present, these are tiny) and the focus is almost exclusively on undergraduate education.

R1 institutions are typically much larger universities with over 5000 students, and some of the flagship public institutions have tens of thousands. They are prestigious in the sense of brand-name recognition. Private R1s tend to be smaller in size and more expensive, but they can be generous with financial aid. Public institutions are more affordable although costs are rising more steeply at the flagships than at the regionals. R1s have a range of graduate programs, and in particular Ph.D. programs in the sciences.

At a SLAC, class sizes are smaller. While you might have a hundred students in general chemistry, some programs split into smaller sections. At my institution, introductory classes are capped at 40, so that’s the typical size of my standard G-Chem class. (Honors classes are capped at 20.) At a large public R1, you might well have 400 students in G-Chem, and there might be several of these large lecture sections being run every year. The size is capped by the limits of available lecture theaters. As a professor, I would much, much rather teach smaller classes where I can learn everyone’s name and even get to know some of my students with their quirks and interests. Pedagogically, I can do different things with a smaller class that would be more challenging with a few hundred students in the same room.

Teaching loads at a SLAC are noticeably higher. Many of the Oberlin Group schools have what is known as a ‘3-2’ load – a faculty member teaches 3 classes one semester, and 2 classes the other semester. I’m limiting the rest of my discussion to tenure-line faculty because there are typically fewer adjunct faculty (with variable loads) at SLACs although those numbers are growing. At an R1, the load might be ‘1-1’ or even ‘1-0’ for tenure-line faculty; and much of the teaching is done by graduate students and adjunct faculty. Graduate students also often do the grading and hold office hours for large introductory level science courses. At a SLAC, the professor does all of this. Hence, a lot of a faculty member’s time goes into teaching at a SLAC.

New faculty members are anxious about trying to get tenure. At an R1, getting tenure is predominantly based on research productivity. This is measured in two main forms: bringing in external grant funding and publishing high profile scientific articles. Giving presentations, serving on discipline-related committees or editorial boards, and other activities that are profile-raising are helpful, but grant funding and getting articles published are the main metrics. Thus, a faculty member’s time and energy is focused on these two activities. To get into the ‘virtuous’ (but vicious) cycle, you need workers to accomplish the research. Thus, one needs to support graduate students and postdoctoral researchers with large, typically million-dollar, grants. Undergraduates (unless very talented and willing to put in lots of time) are typically not as productive, since they have classes and other commitments. As a faculty member, much of your time is spent writing (grant proposals and articles).

In contrast, undergraduate research is a selling point of SLACs with strong science programs. As a faculty member, much of my time is spent directly mentoring the students through the research process. (I get to know my research students well!) Yes, I do spend time writing proposals and getting articles published, but it’s a much smaller slice of my schedule. I’m also directly involved in generating research results; that ‘extra’ bit of work on my part is often needed to turn an undergraduate research project into a publishable article for a ‘good’ or at least ‘decent’ journal. I am able to small chunks of research and writing during the semester, but it mostly gets done during the summer. When classes are in session during the semester, that takes the lion’s share of my time.

At a SLAC, you have to be a good-to-excellent teacher to achieve tenure. You also have to show some research productivity (grants and publications) although not at the quantitative levels of an R1. At a smaller school, service to the department and university is also an important component. (Someone has to serve on committees and help with administrative work!) In recent years, service expectations have been lowered for junior faculty as research expectations have crept higher. At an R1, as long as you’re not a poor teacher or generating lots of student complaints, that’s typically good enough to satisfy the teaching part of the portfolio in the sciences since research has become the all-important criterion. Junior faculty members are typically ‘shielded’ from service so they can spend more time beefing up their research portfolio.

Academic advising is an important part of being a SLAC faculty member. You meet with your advisees regularly, help them navigate their academic lives through college, discuss career and post-college plans, and get to know them as individuals. I very much enjoy this part of my job, although it does take time. At an R1, an undergraduate may get little to no advising from a tenure-line faculty member unless you were working in the lab. Many schools have moved towards having professional career academic advisers who may meet with a few hundred students per semester. Forging connections is so much harder.

In a small department, the faculty and staff all know each other and (hopefully) work well together. Much of the equipment and responsibilities are shared because that’s the only way to function effectively without millions of dollars of grant funding to pay for equipment and personnel in one’s own research lab. Teaching and research are blended, with plenty of overlap at a SLAC, while they are generally considered very separate activities at an R1. Thus, the division of labor and resources differs significantly between the two types of institutions. In an R1 the faculty may not all know each other, or the staff, or the student majors. At a SLAC, all the student majors are known personally by one or more faculty members.

From conversations with colleagues in areas outside the natural sciences, the differences between being a faculty member at a SLAC versus an R1 are not as stark, although there are differences. Although not very common, faculty occasionally move between the two types of institutions in the humanities and the social sciences. Such moves are very rare in the sciences. At the outset, the stark differences for faculty in the natural sciences means that someone applying for a faculty position needs to clearly understand these differences and tailor the application accordingly. Having sat on many hiring committees, I can say that not understanding what it means to be at a liberal arts college in the sciences pretty much disqualifies a candidate given the strong applicant pools. As a postdoctoral research fellow, I made a clear choice to work at a liberal arts college and not at a research institution because I wanted undergraduate education (in its various forms) to be my focus.

A final cautionary note: As interest in a liberal arts education increases outside the U.S., and we see new institutions and partnerships sprouting, I have noticed a lack of understanding as to what a liberal arts college is – even as efforts are being made to replicate some of its aspects. A liberal education or liberal arts curriculum is not the same as a liberal arts college. Yes, liberal arts colleges by definition subscribe to a liberal arts curriculum, but many universities in the U.S. have such curricula but would not be anything like a liberal arts college, from the experience of faculty, staff and students. However, the two have often been conflated, and not recognizing the stark difference between institutional types (irrespective of curricula) has led, in my opinion, to challenges and problems in science departments trying to navigate the morass and confusion of differing ideas and unrealistic expectations.

Revisiting Ungrading

I’ve made several attempts to reduce the emphasis on grades in my classes. For example, I have written an open letter to students signing up for my class to explain what I am doing and why. However, it’s not enough to present this in a class syllabus and discuss any issues of concern on the first day of class. I’ve learned, through trial and error, that the what and the why must be constantly reinforced. Students often have a fixed notion of how grades work in a science class, not through any fault of their own – the successful ones have adapted well to the ‘system’. Therefore, changing the way things work introduces some anxiety, especially among students who have ‘done well’ in previous classes.

There’s nothing wrong with wanting to know how you’re doing in a class as a student. It’s an important skill to ascertain what you don’t know and what else you need to learn. The trouble is that students easily fixate on the grade and don’t think about the learning. The grade is a quick-and-easy measure, a proxy of sorts – an A means you’re doing well, a C perhaps not so great. Part of the problem is that the purpose of grades has been conflated in the two types of assessments used to gauge ‘learning’: formative assessment and summative assessment. It’s not just students who confuse the two; teachers do the same. Summative grades were used early in the meat packing industry. Over time, they have completely pervaded the education industry (and yes, it’s an industry).

If you regularly read higher education news, you’ve likely heard of the Ungrading movement among college professors. For example, when ‘grading’ a paper, a professor writes comments without providing a score or letter grade. In the paper drafting process, students ideally use those early comments to improve. The draft is not a ‘final’ work that is assessed summatively; the comments provide formative assessment. The professor might assign a tentative grade in his or her gradebook without telling the student what it is, and revise that grade after assessing the final paper. But when you’re grading a large stack of papers, and you want to motivate students not to turn in complete rubbish in the first draft, you might explicitly assign a portion of the student’s final grade to the quality of the draft. This fear-motivation can be well-meaning and practical, but it blurs the distinction between summative and formative assessment, and furthermore reinforces the students unhealthy focus on ‘the grade’.

More recent versions of Ungrading go a step further. Students are asked to reflect on their learning and suggest what their final grade should be at the end of the class, based on their learning progress and their achievements. There are caveats, though. The professor still ‘controls’ the grade and may disagree with the student’s assessment, thus providing some standard of summative assessment. While most examples have come from the Humanities, where students typically write papers and such assessments are often assumed to carry a higher degree of ‘subjectivity’, there are recent movements in the sciences using the same approach. Including, yes, organic chemistry.

My recent forays into Ungrading focus on a cleaner separation between formative and summative assessment. I’ve done this by asking students to participate in annotated self-grading. In G-Chem, I’ve introduced this for midterm exams (I typically give three to four per semester). In P-Chem, the students annotate their problem sets. My approach thus far does not save me any time in G-Chem, but I have noticed a significant reduction in my ‘grading’ time for P-Chem when I look over the problem sets. In both classes, if students make a good faith effort to follow my instructions, I give them full credit, i.e., they are not penalized. The whole point of this approach is to encourage the students to reflect on what they know and what they don’t know – so they can better target their learning. Is it working? Partially, I would say. I don’t yet have enough data, but I suspect that it doesn’t work so well for struggling students at the low end. My approach needs refining.

Last fall, I introduced a ‘replace your lowest exam grade’ option in my Quantum class. Students could opt to submit a reflection on what they had learned related to the material that was tested on the exam they wanted to ‘replace’. I had what I thought were clear guidelines in my syllabus, along with an exhortation that the reflection needed to be substantial and would benefit from weekly journaling BEFORE even taking the exam. I hoped that students would, in the process of reflecting, actually learn the material in greater depth and then actually do better on the exams. This didn’t happen, of course. Again, my approach needs refining.

So, I’m thinking of an approach that combines my reflection-journaling idea with the ‘suggest your grade’ approach my upcoming special topics course a year from now (because I’m on sabbatical!). I think the special topics course lends itself well since the class will primarily consist of reading and discussing primary literature. Students also participate in leading some of those discussions and coming up with guiding questions for papers that we read. There will be a final project culminating in a written research paper and an oral presentation. I am planning to thread reflection-journaling assignments throughout the semester that will culminate in a final reflection where each student proposes his or her overall course grade. I haven’t worked out all the details yet, but I’m excited about the prospects of my Ungrading approach for this course! I’m sure students will be anxious and I will need to figure out how to allay their concerns while promoting the benefits of my approach.

Minimizing Methane: Cow Version

One piece of advice I give to my students every semester in G-Chem is to be wary of lighting a cigarette in a field of cows. Especially if all their butts are facing you. My students are unlikely to face such a scenario in this day and age, but the trope always elicits laughter every time I’ve used it.

The oxidation of methane (CH₄) via combustion is a standard example in G-Chem. I use it in the first semester when students learn stoichiometry and how to balance chemical equations, but it also gives them practice drawing out the Lewis structures of all reactants and products. They can track which bonds are being made and broken in the chemical reaction, and given average bond energies make an estimate of the energy released in the reaction. The energy released is substantial per unit mass, and this makes methane (the main component of natural gas) an excellent fuel. In the second semester, we examine fuel efficiency in detail when we cover thermodynamics and kinetics, and methane combustion is compared to other fuels.

The classroom I teach in has a demonstration bench with a sink and a gas line. Students in G-Chem lecture are also co-enrolled in the laboratory, and they have covered lab safety issues and what to do when you smell ‘gas’. (Methane has no odor, but an additive is added so that human noses can quickly detect a gas leak.) There are enough cow fart jokes around that students typically know that cows release methane, and I probably discuss ‘natural’ gas in a humorous light (all puns intended). In class, I mimic turning on the gas line to prepare the students for the scenario; they should prepare to run away! (We also discuss why there is no explosion when gas leaks into a room full of oxygen until a spark lights things up! I've mentioned this in a previous blog post that discussed the possible origin of dragon fire.)

In any case, this week I was happy to read the following short article in Chemical & Engineering News (C&EN) on minimizing methane from cattle. It gives me some new factoids to add to my cow fart advice, but allows me to introduce more chemistry! Here’s what I learned: The annual methane release from a cow is typically 70-120 kg but the majority is released by burping, not farting. I’ll have to change the story I tell my students so they are aware that cows facing you are potentially more problematic. Also, about 7.5% of global greenhouse emissions comes from cows. That’s a big number! The hero of the story however is a cattle-feed additive with the active ingredient 3-nitrooxypropanol (3-NOP, but I will call it NOP for short). The picture below comes from the C&EN article where I have overlaid the line structure of NOP. The cow’s mouth is appropriately wide open!

The claim made by DSM, the company trying to add NOP to cattle feed, is that “a quarter teaspoon per cow per day is enough to inhibit the formation for methyl-coenzyme M reductase, an enzyme used by methane-generating microbes in a cow’s digestive system.” Tests on 48 cows over 12 weeks saw a 30% reduction in methane emissions. As a bonus, NOP also reduces the cost of feed (by 3-5%) because cows use less energy when they generate less methane, thus the cows don’t consume as much feed. Sounds pretty good, doesn’t it? It will be interesting to see how NOP competes with other ‘natural’ methane-minimizing additives (flax seed, linseed, garlic and citrus extracts), assuming its use is approved by regulators.

I might also be able to bring up this example when discussing climate change. Who knew that cow emissions could be so interesting!