Planning update

So, I've updated the Completed, Current, and Upcoming courses pages.

Not much to say, except it's noteworthy that my upcoming planning is heavily Courserized. My beginnings in the MOOC world were through edX, and I still prefer its linear approach (resources are arranged in chronological order, while on Coursera they're arranged by resource type) and richer exercise types (formula entry, molecule editors, etc. go well beyond Coursera's simple quizzes and peer-reviewed essays). But while the first courses I've taken from Coursera have been a bit disappointing, I've take a couple great ones since, and after a while I've warmed up to the platform (and its much, much better forum software). Although in reality, what's importance is the vastness and breadth of the course catalog: Coursera's is what, five or six times larger than edX? So, being interested in a somewhat limited range of topics (bio, data analysis, comp. sci, with occasional forays into economics) I'm finding that I've, well, “exhausted the edX catalog” isn't the right term, but conveys the right impression − what I mean is that there are many more courses on Coursera that I'm interested in, and that's purely an artefact of numbers.

Finished: Epigenetic Control of Gene Expression

I'll write something later, but just a quick note: I passed Epigenetic Control of Gene Expression from Melbourne University through Coursera. I guess it's the highest-level biology course I've taken so far (the professor says it's about either final-year undergraduate or first-year graduate). My final score should be 97% or thereabouts, so I guess that means I mastered the content pretty well.

So, I…

• … upgraded to Verified student in MIT's Quantitative Biology Workshop (hints have been dropped that MIT is considering a Biology XSeries, and that taking verified certs now means taking a head start),
• … registered for TU Delft's Technology for Bio-Based Products,
• … and for the University of Illinois' Emergence of Life,
• … and for Duke's Introductory Human Physiology,
• … and for the University of Alberta's Dinosaur Paleobiology.

And so, my schedule is full until November, approximately.

Why I dislike "guided discussion"

Some course designers feel a need to ask students to discuss course-related topics. For instance, at the end of a segment, they'll write “What do you think about genetics? Please discuss this in the forums.”

This kind of open-ended discussion works fine in smaller, physical classrooms, where only one person may speak at a time and everyone must listen to everything that is being said. In such situations, a real discussion may happen and insights may percolate.

In online classes of hundreds of thousands, you get…

… noise. Hundreds of single-message threads, never leaving the realm of platitude.

Course designers: please, please, please do not give in to the temptation of using “guided discussion”. Given the present state of technology, it only renders the whole course forum useless.

I can MongoDB

And so I passed MongoDB's Advanced Operations and Deployment course. My score is 100%, but in fairness it doesn't mean much: it's one of these courses where either you get it, or you don't; there's little place for a middle ground.

Still, it's nice to ace a course.

On the radar

Here are a few courses I've noticed but haven't committed to yet (and couldn't possibly commit to all):

edX

Technology for Bio-based products from TU Delft

Improving Global Health from Harvard

Behavioural Medicine: a key to better health from the Karolinska Institute

English Grammar and Style from the University of Queensland

Shakespeare: On the page and in performance from Wellesley

Transforming Business, Society and Self from MIT (because this sounds weird yet comes from MIT)

Coursera

Programmed cell death from LMU München

Emergence of Life from U. Illinois

Introductory human physiology from Duke University

Introduction to Systems Biology from Mount Sinai (I tried that one earlier in the year, and dropped it because I was too tired, too busy, and the course was too demanding. I am still interested in the whole Systems Biology specialization, but am more than a bit put off by the, erm, distinctive teaching style of this first course.)

Dinosaur Paleobiology from U. Alberta (yeah, in my heart I'm still a little boy with a fascination for ancient beasts reptiles creatures non-avian dinosaurs.)

Drug discovery, development and commercialization from UC San Diego

Social Network Analysis from U. Michigan (which uses Logo! Turtles all the way!!)

Climate Change from the University of Melbourne

FutureLearn

Hadrian's Wall: Life on the Roman Frontier from Newcastle University (where else?)




I will not, obviously, take all of them (the number of hours in a given week being, sadly, of limited elasticity). The ones I'm not likely to take, right now, are Biobased products at Delft and Emergence of Life at UI. I'll probably take one of the other of Physiology and Dinosaurs, too.

End-of-course blues

So… I'm in the final throes of two courses now: Epigenetic Control of Gene Expression from Melbourne University (Coursera), and MongoDB Advanced Deployment and Operation (MongoDB). In addition, ASTRO1 from ANU is also officially closing in a few days − I already have my final score for ASTRO1, in contrast to the other two, and anyway ASTRO2 is starting in a couple of weeks.

I always feel a bit bluesy when a course ends. I guess that's because, well, it's the end: all of the time and effort you've invested in the course have born their primary fruit (some additional knowledge or skills, maybe a bit of paper to prove it), but − unlike in a “real” course − you don't go on hanging out with classmates, you won't be bumping into the professors in hallways, etc.; finishing each course is in that respect more like graduating: the end of something.

Then of course, when I've made time for a 7, 10, or 15-week course in my schedule, when the course ends, there's a hole to fill in − things can feel a bit empty at first.

So, onwards! To recap, in the first half of this year I've completed the following courses:

• Economics and Calculus at Caltech
• Statistics part 1 and 2 at Berkeley
• Immunology part 1 at Rice
• Analytics at MIT
• Bioinformatics at Peking University
• Astrophysics part 1 at ANU
• Diabetology at Copenhagen University
• Epigenetics at Melbourne University
• MongoDB administration (both courses) at MongoDB

(Okay, so Epigenetics isn't quite over − I still have to grade my peers' assignments − and it is possible that I utterly failed the final essay and therefore the course; but that's very unlikely.)

Besides, I am now enrolled in the following:

• Statistics part 3 at Berkeley
• Genomic Medicine at Georgetown
• Quantitative Biology at MIT

I may drop the Genomic Medicine course − it's interesting, but it feels way too basic, especially after the rather challenging Diabetes and Epigenetics courses. Of course, I have now a bit of time on my hands, so may stick to it just to get busy.

Later on, I am pencilled in a few additional courses: ASTRO2 (Exoplanets) from ANU, two analytics/statistics courses (one focusing on healthcare data from the Karolinska, and a more general one from Duke), a couple of medical courses (Immunology Part 2 from Rice, and Anatomy from Harvard), and a discussion about Scottish Independence from the University of Edinburgh (obviously centered on the referendum date so as to discuss the issues as well as the aftermath).

So − that's eight more courses I'm either currently doing or scheduled to take before the year's out, bringing the grand total of 2014 to a whopping 19. I am quite certain I'll find an additional course to take it up to 20, which when you add the 5 certificates I got in 2013 makes for a not completely awful curriculum.

Worth Sticking To? Genomic Medicine Gets Personal, Georgetown University

We're now in Week Two of this course, and getting to grips with the actual content. As a reminder, this course from Georgetown University is about the transformative aspects of “genomics” as a whole with regards to the theory and practice of medicine.

In other words, while I've been studying basic biology, bioinformatics, epigenetics, etc., from a purely intellectual point of view, this course takes a different approach, grounded on the experiences of doctors counseling actual patients. Which is all and well, only… well, the actual science is too basic. I am not certain I can actually be bothered to listen to people explain that there are 22 pairs of autosomes, or what a mutation is. So while the point of view is certainly interesting, at this stage I am not quite certain it is enough to keep me interested. We'll see; but in the meantime, this course is in the uncomfortable position of the one I am most likely to drop.

Starting: 7.QBWx Quantitative Biology Workshop

Let's see… 7.00x Introduction to Biology got me hooked on to MOOCs. I've been assiduously taking courses in biology / medicine ever since, as well as computing courses and bioinformatics ones. So, when 7.QBWx was announced − a hands-on, workshop-oriented, introduction to the realm of “quantitative biology” (i.e. bridging the gap between data scientists and biologists), from the great people at MIT (were it not for the unfortunate Social Physics Buy-My-Book-Please-But-Let's-Pretend-It's-A-Course, I'd believe that MIT is the place where the exceptionally deserving go to when they die) − well, what could I do − except end this rambling sentence?

While the official start date was yesterday, the course hasn't really started − we've only been given a handful of introductory tasks to complete, basically checking that we could install all the necessary software on our computers. The real stuff begins next week. In the meantime, my freshly-updated Linux Mint is now equipped with R, Canopy Python (on top of system Python), Octave, and PyMOL.

I am slightly apprehensive: how can a 6-week workshop covering so many different tools and languages actually achieve any significant learning outcomes? The only way I can see is to make the course pretty hard − I'm not worried: by now I have a decent enough biology background, and of course I've been a professional programmer for 15 years, so I doubt I'll struggle in this course − which is my worry, really: if I don't struggle, how am I going to actually learn stuff?

Well, we'll see. I am fairly confident the great people at MIT (where the exceptionally deserving, etc.) have assembled a challenging, great course, and am eager to see what it comes to.

On a side note, a staff member (pseudonymed TurtlesAllTheWayDown, which − as a long-time Pratchett fan − I find absolutely wonderful) has admitted that an MIT Biology XSeries is at least being discussed. I am incredibly excited about this − and will therefore probably upgrade to Verified Certificate after next week, if it appears the course is really worthwhile, so as to make it count towards the future, possible XSeries.

Reading scientific papers

The ubiquitin multi-motif protein UHRF1 is a central player in targeting repressive chromatin marks. It contains a SRA domain, which binds to hemimethylated DNA, a Tudor domain binding to methylated H3 (H3K9me3) and a PHD finger interacting with an unmodified arginine residue within H3 (H3R2) [72–74]. Furthermore, UHRF1 interacts with DNMTs, G9a and HDAC1 and thereby unites various enzymes that can provide a repressive chromatin environment [75–77]. Interestingly, UHRF1 also recruits the H2AK5 actetyltransferase TiP60 thus integrating a multitude of different epigenetic signals [78]. A further example for the link between DNA methylation and histone modifications represent methyl C binding proteins such as MeCP2, which interact with co-repressor complexes including HDACs and HMTs [79,80]. Interestingly, a recent report shows that components of the piRNA pathway are required to target de novo DNA methylation to an imprinted region of the mouse genome implicating that selective methylation of imprinted regions can be regulated by non-coding piRNAs [81].

Imagine 21 pages of this. Okay, 9, if you discount the bibliography. That's the review article we're supposed to read (and understand) for the final part of the Epigenetic Control of Gene Expression MOOC at Melbourne University.

Well… I'm finding it a lot of work to read the text, parse the sentences, expand the abbreviations, and finally integrate it into a whole. I suppose it's mostly due to a lack of training − if I were to read that kind of paper on a daily basis, I guess I would acquire some paper-reading skills. That, or my brain would overheat and melt (and believe me, it's embarrassing to have to mop up your own liquefied brain matter that's dripped through your ear ducts to the carpet.)

I guess I'll have deserved that certificate.