This month on the PDB: December

Hi everybody! 162 structures were released from November 28th to December 5th, ranging from the typical Homo sapiens proteins to the zesty proteins of the Asian rice. Without further ado, let us take some time to peruse this newly released selection of never-before seen macromolecule structures!

  1. HSP90 WITH [sic] indazole derivative, by Graedler, U., Amaral, M., & Schuetz, D.. You know how part of what makes the PDB exciting is that the PDB releases never-before-seen-structures? Well this is not one such case, but just like the lysozymes of last week, this title is notable in that it is actually the name of not one, but SIX separate releases for this week. Hsp90 (heat shock protein 90) is a chaperone protein that promotes the proper folding, stabilization, and activation of a client polypeptide in an ATP-dependent manner, commonly as a dimer. In teaching materials, Hsp90 usually looks like some weird ellipses lumped together to form some two-pronged rabbit head-looking thing, but now you can see what it really looks like, in six slightly different conformations! Indazole derivatives are Hsp90 inhibitors, and these monomeric structures of human Hsp90 are each binding a slightly different indazole derivative. Why are they monomeric and what’s with all these different indazole derivatives? Well, the paper hasn’t been published yet so you’ll have to keep on guessing for a while, but until then, you can entertain yourself by looking at the structures. The PDB accession codes are 5LNY, 5LNZ, 5LO0, 5LO1, 5LO5, and 5LO6. The following figure includes an approximate surface density of the protein, which is why it probably looks a little strange. You can see there are slight differences in conformation between the structures, but for the most part they look decently similar.3.1
  2. Crystal structure of Os79 from O. sativa in complex with UDP, by Wetterhorn, K.M., Gabardi, K., Michlmayr, H., Malachova, A., Busman, M., McCormick, S.P., Berthiller, F., Adam, G., and Rayment, I. This is actually a generalization of the titles of four different structures, all of Os79 from O. sativa in complex with UDP, with and without different mutations and some with different sugar moieties. Oryza sativa, the humble Asian rice, one of the most essential cereal crops to society, is susceptible to head blight infections caused by fungi in the Fusarium genus that affects many other cereal crops as well. Trichothecene toxins are a family of toxins that are responsible for the virulence of Fusarium head blight, and toxic to humans and livestock as well as plants since it inhibits protein synthesis in the eukaryotic ribosome. Os79 is a UDP-glycosyltransferase: it adds a glycosyl moiety to a substrate using a co-substrate such as a UDP-glucose (a glucose bound to a uridine diphosphate). Glycosylating trichothecenes reduces its bioavailability and toxicity, so it would be super rad to engineer a protein that could easily glycosylate these toxins and prevent humanity from losing a lot of crops, some livestock, and a few humans every year. That’s a lot of lives saved! Os79 seems to be a good candidate for this ^type of protein engineering, but this goal is still in the fledgling stages as there are a lot of different trichothecenes and only one Os79 (Remember that enzyme-substrate specificity which we thought made enzymes so cool? Well now it’s sort of biting us in the butt. Enzymes are still cool though, obviously). The good news is that the group that released these structures also did some structural and activity analyses and determined what parts of the structure controlled its specificity. You can read all about it here: The PDB accession codes are 6BK0, 6BK1, 6BK2, and 6BK3.3.2

Someone requested the experimental method metrics of the PDB releases, so here they are: out of 162 structures released this week, 146 were obtained through x-ray diffraction. 4 structures were obtained using cryo-electron microscopy, the up-and-coming technology in the world of structural biology (the 2017 Nobel Prize in Chemistry was awarded to the developers of cryo-EM), a lower number than the usual. This week, there are a whopping 12 structures obtained through solution microscopy. What are these mysterious 12 structures? The 4 cryo-EM structures? The 144 x-ray diffraction structures? (You already saw 2 of them.) Go check them out yourself on this week’s PDB release! You can find it here: Happy lurking!


This week on the PDB: November 24th – November 30th

Welcome back to another week of “This week on the PDB”, where I discuss  you a very small section of new, never-before-seen protein structures uploaded to the Protein Data Bank, because proteins just keep on getting discovered.

  1. Lysozyme is an enzyme essential to our immune system that destroys bacterial cell walls. It is ubiquitous in the body: tears, mucus, blood, you name it, you can probably find lysozyme there. Lysozyme also holds a special place in the PDB, as not only was it the FIRST structure to be deposited in the PDB, back in the day when the whole PDB consisted of only 7 structures, but it is also the protein with the MOST different structures deposited, largely owing to a series of experiments conducted Brian Matthews where he made various (and by “various”, I mean “hundreds of”) mutations to lysozyme. Lysozyme is also one of the most consistently crystallisable proteins and has consequently been used to study the protein crystallization process. This week, lysozyme has once again reared its head on the PDB, where Hosur et al. deposited four new structures of lysozyme, at different time increments in the guanidine hydrochloride and glycerol soaking process. The picture below is of the four structures and the ligands aligned with each other. The PDB accession codes are 5H6A, 5H6C, 5H6D, and 5H6E. The associated article has not been published yet, so it the purpose of this structure is unclear, but be sure to check out all the other hundreds of lysozyme structures on the PDB, which can be accessed here.


    Hen Egg White Lysozyme native crystals soaked in precipitant solution containing 2.5 M guanidine hydrochloride and 25% glycerol.

  2. PolyA polymerase module of the cleavage and polyadenylation factor (CPF) from Saccharomyces cerevisiae. In order for a freshly transcribed piece of RNA to mature into a useful piece of mRNA for translation, its 3’ end must be cleaved, and a string of adenines added to form a polyA tail. CFP mediates this whole process, making it one of the MVPs of mRNA processing, but for some reason, we don’t really know how this amazing complex of proteins so important to our existence assembles itself (shocking, I know. I rank it high in my list of “why don’t we know this” along with the polymerases involved in DNA replication, where there has been ambiguity for quite some time now). At least, until Casañal and Kumar et al. solved the structure of nuclease, polymerase, and phosphatase modules of the CPF in the common baker’s yeast, S. cerevisiae, using electron microscopy with a resolution of 3.55Å. You can read all about the structural features, such as the four beta propellers (ooooohhh), of the yeast—a great model eukaryote—CPF here. The PDB accession code is 6EOJ.


    Cleavage and polyadenylation factor (CPF) from Saccharomyces cerevisiae

  3. Crystal Structure of pro-TGF-beta 1. This structure of the pro-transforming growth factor-beta 1 from the wild boar was obtained through x-ray diffraction by Zhao, B., Xu, S., Dong, X., Lu, C., Springer, T.A. with a resolution of 2.9Å. TGF-beta 1 mediates many cellular functions, including growth, division and proliferation, differentiation, and apoptosis (programmed cell death). It is also important for the immune system, where it can interfere with other cytokines involved in the cellular immune response. In general, it is a very important cellular signalling protein, specifically a cytokine. It also has an implication with tumor development, and consequently holds relevance to cancer (as we all know, anything with connection to cancer is a hotbed of potential biomedical research). This structure has a swap between the N-terminal prodomain and the C-terminal GF domain, which appears to affect how the proteins assemble with each other. The associated article is heavy on the structural biology and lighter in clinical implications, but can be read here. The PDB accession code is 5VQF.2.4

The previous week, a total of 220 structures were released on the PDB. While many of them were simply different resolutions of the same protein (not to mention those four lysozyme structures), there are still a plethora of structures released every week. This just goes to show the expansiveness and diversity in the realm of macromolecules. You can find all the PDB releases this week here: See you next week!



This week on the PDB: November 7th – November 13th

Welcome back to “This Week on the PDB” for the week of Nov. 7th! This week on the PDB, 133 new structures were released, providing 3D visuals for macromolecules found in species ranging from the common human to the HIV virus. Here are some highlights:

  1. The solution NMR* structure of the Membrane Associated Segment of HIV-1 gp41 Cytoplasmic Tail was released by Murphy, Samal, Saad, and Vlach. This HIV-1 domain is important in mediating the recruitment and incorporation of the viral envelope protein (Env) into the virion. The full paper can be found in the November 7 issue of Structure, DOI: The PDB accession code for this macromolecule is 5VWL.
  2. Human glutathione s-transferase Mu2 complexed with BDEA in a monoclinic crystal form was released by Zhang et al. The structure was obtained through x-ray crystallography and had a resolution of 1.6 Å. The asymmetric unit consists of two chains, A and B. Glutathione s-transferases (GST) make up a family of proteins that add glutathione to proteins, which aids in their regulation. GST is important in cancer research, and deviations from normal GST structure appears to correlate with increased susceptibility in certain cancers such as lung, prostate, and colorectal. Though glutathione s-transferases (GST) are generally well characterized, and many other GST structures can be found on the PDB, this appears to be the first structure of GST M2. Additionally, this GST is complexed with an inhibitor, and glutathione, giving further insight into how this GST M2 interacts. The article for this structure has yet to be published, but for now you can look at the structure by searching up its PDB accession code, 5HWL.


  1. Cryo-EM structure of human insulin degrading enzyme was released by Liang et al. The resolution is 6.5 Å and the asymmetric unit contains two chains. Insulin degrading enzyme, as the name suggests, is involved in the degradation of insulin in the cell, as well as IAPP, glucagon, bradykinin, kallidin and other polypeptides, which allows insulin degrading enzyme to affect certain intercellular signalling pathways. Insulin degrading enzyme also breaks down amyloid formed by APP and IAPP, which might have implications in neurology. An article has yet to be published, but for now you can see it on the PDB, with accession code 6B7Y.


  1. Human Apo-TRPML3 channel at pH 4.8 was released by Zhou, Li, Su et al. The technique used was cryo-electron microscopy and the resolution is at 4.65 Å. TRPML3 channels are mostly found on the endolysosomes and are critical to the endocytic pathway and consequently cell signalling. This apo (unbound) structure at pH 4.8 is very different from the apo structure at the physiological pH of 7.4, and the lower pH appears to inhibit channel activity. Malfunctioning TRPML3 channels can cause deafness and interfere with proper pigmentation in mice, but does not appear to correlate with any human diseases yet. However, malfunction of the closely related TRPML1 channels in humans cause a severe lysosomal storage disease, mucolipidosis type iv. The paper was published November 6 online in Nature Structural & Molecular Biology, and can be found here: Its PDB accession code is 6AYG.


This just scratches the surface of this week’s released on the PDB. If you are interested in seeing what the other 129 structures are, you can find them at Happy viewing, and come back next week for more structure releases!

*Solution NMR is a technique where the protein is suspended in a buffered solution and multidimensional nuclear magnetic resonance is applied to the solution. Radioisotope-labelled proteins are also used during the process. Through resonance assignment and by creating different restraints for parameters such as distance and bond angles, a 3D structure can eventually be generated. Solution NMR doesn’t require a crystal to collect data, unlike the case of x-ray crystallography, and also has the added benefit of being able to provide information on the dynamics of the protein, but is limited to small-sized proteins.


This Week on the PDB: An Introduction

Protein functions are largely dependent on their 3D structure, but where can we find these structures, and more importantly, manipulate them? The answer: The Protein Data Bank (PDB).

The original PDB was established in 1971, and contained a total of seven structures. Since then, it has grown to contain more than 133759 structures. Though the name suggests that only proteins are covered, the scope off the PDB is much wider than just polypeptides. Ligand-bound proteins, nucleic acid-bound proteins, viruses, ribosomes, and just about any other kind of macromolecules can be found on the PDB.

The PDB has a built-in 3D structure viewer accessible through a browser. Using the 3D viewer, you can spin the structure in any direction to get a full 360o view, zoom in, and zoom out. You can also view structures in different styles such as cartoon or ball-and-stick, and you can see the plot of the surface as well. Other features include the amino acid sequence of the protein and seeing how specific residues interact with ligands.


An alignment made in PyMol, with sequence view turned on.

The PDB’s built-in viewer has all the features needed for a cursory glimpse of the structure, but lacks features required to do in depth structural analyses. As such, most people tend to download the .pdb file of the structure and view it in another program. The PDB is so ubiquitous in structural biology that structural analysis programs such as PyMol can fetch PDB files just using the 4-digit alphanumeric code associated with a given protein, called the PDB accession code.

Most of the structures on the PDB are obtained through x-ray diffraction, but other techniques, such as electron microscopy, can be used to solve for the structures. There is also a database specifically for structures solved from electron microscopy called the EMDataBank, which is easily accessible from the PDB. The reverse also applies.

Though the PDB contains an immense database of protein and other macromolecular structures, there are still an extraordinary number of macromolecules that have yet to be catalogued. Consequently, the PDB updates once a week, adding newly solved, never-before seen structures.

PDB intro

The main page of the Protein Data Bank.

Now that you have been primed on what the PDB is and how it can be used, I would like to introduce a new column on The Abstract: “This Week on the PDB”! Every week we will present to you the new structures in the PDB, and a short summary of their purpose. Together, let us explore the cutting-edge science of structural biology.

Check out the PDB here:

What is Infinity?

To read more by Shunsuke:

“I love you to infinity”, “I love you to infinity plus one”, “I love you to infinity plus infinity”, …

Let’s look at this mathematically, together, as you were just about to do on your own.

Mathematicians are to mathematics as geologists are to rocks.

Not many people gain pleasure from observing an ordinary rock, just like nobody ever gets excited when seeing an integral. Rocks have been in our hands as tools for a very long time, and we have developed many great things from them. Similarly, mathematics has been the flashlight we’ve used to discover new things: from the Pythagorean theorem to Einstein’s theory of general relativity. But it’s not mathematics that’s fascinating, but the results that we find beauty in. While having said that, geologists look very carefully at rocks and listen to the stories that they have to tell.

So let’s listen carefully to the sweet sounds of whisper of mathematics – to infinity and beyond.

Yes, infinity – the idea of infinitude is as beautiful as it is vast. Let’s put aside the “you’re infinitely beautiful,” “no, you’re infinity plus infinity times more beautiful,” “no no, you’re infinity multiplied by infinity times more beautiful” he-said-she-said – neither is complimenting the other any more than what they just said. As heartbreaking as it was when Andy had to leave for college in Toy Story 3, there are so far only two (sadly not infinite) types of infinities; namely, countable and uncountable infinities.

Countable infinity is a set of numbers that if you lived to be infinity years old, you can count all the numbers in the set. For example, the set of odd numbers {1, 3, 5, …}. If you count an odd number everyday for infinite number of days, you can count all odd the numbers. Let’s take a closer look at this after we define “uncountable infinity”.
Uncountable infinity is a set of numbers that even if you lived to be infinity years old, there’s no way that you can count all the numbers in the set. An example is the set of real numbers which is all the numbers that can be represented by decimals. You can spend an infinite amount of time counting all the numbers between 0 and 1 but you’ll still have the numbers between 1 and 2, 2 and 3, and infinitely many remaining intervals.

So, back to countable infinity. Here is a question: which set of numbers do you think is bigger, the set of prime numbers {1, 2, 3, 5, 7, 11, …} or natural numbers {1, 2, 3, 4, 5, …)? Euclid proved that there are an infinite number of prime numbers, which is a cool little fact, but would you believe me if I told you that there are exactly the same number of prime numbers as natural numbers?

Let’s say that there is a set of boys each representing a natural number and, likewise, a set of girls each representing a prime number.  We are going to form couples to have them go on a blind date. If there are more boys than girls – in other words, more natural numbers than prime numbers – we’ll know that by seeing that at one point we’ve run out of all the girls and we’ll have a (fairly sad) situation with only boys. But that’s not the case, since even when 20 couples or 100 couples or a million “couples” are matched, there will still be an infinite number of boys and girls that can be happily matched.
This is called the one-to-one correspondence: we will never find a boy or a girl – a natural number or a prime number – who cannot “find a date”. That may sound unrealistically perfect for the real world, of course,  but the world of mathematics is as perfect as it gets.
So, to recap: infinity = infinity + 1 = infinity + infinity = infinity × infinity. Those lovebirds were just repeating the same thing to each other [1].

Another fun fact is that the number of rational numbers – the set of all the numbers which can be written as fraction (e.g. 1/2, 2/3, 11/4) – is countable infinity as well. This means that there is the same number of rational numbers as prime numbers or even numbers.

Now that we have a little taste of what infinity is, do you think there is another infinity that we don’t know of? Maybe something that’s in between uncountable and countable infinity? I have no idea what that infinity might look like, but if you have an idea, perhaps this branch of mathematics – called number theory – might be for you.


[1] Assuming that these infinities being referred to are countable infinity.

Breaking memories: one data point at a time

Breaking Memories_2By Andrea Weckman | Illustration by Wanying Zhang

I began my first undergraduate research experience in the summer after my first year with a glorified impression of what it was going to be like. As I walked into the lab on the first day, I said to myself:

Andrea, this is your ticket to the Nobel Prize.

I think you’ll find that several things are a little off about this sequence of events. First, and most notably, I was talking to myself. (But every research scientist does, right? It’s normal, right?)

Secondly, at age 18, I was telling myself I would win a Nobel Prize: the gold medal of lifetime achievement, the ultimate symbol of scientific prestige. It didn’t even stop there — I’ve played with the idea of being the next Marie Curie, tucking two Nobel Prizes under my belt. Was I overestimating my role as a first-year undergraduate researcher in a small lab in Waterloo, Ontario?  Maybe. But a girl can dream.

This past summer, as I walked into St. Michael’s Hospital in Toronto to start another unpaid, 9-5 summer research experience, I said to myself:


Andrea, mark this day on your calendar as your first step towards saving the world. 

Again, I was glorifying my role – telling myself I would save the world. (Side note: when I say this to myself, I envision myself as a scientific Superman). But as I progressed through my first experience and then moved on to my second and third, rather than having my notions of grandeur squashed by reality, I came to realize that this delusion is necessary for maintaining your sanity in the world of undergraduate research. As you dig through piles and piles of data and run the same tests countless times, you need that sense of a greater purpose for your research. Your biggest motivation is the possibility of your efforts leaving a mark.

Now, at my current research position, as I make what seems like millions of histograms and residual plots and run different SPSS statistical analyses on what seems like billions of data points, the possibility of making a difference is what drives me.

I’ve met many people in my current research experience, all suffering from posttraumatic stress disorder (PTSD). I’ve encountered people whose memories keep them from sleeping at night, who startle when somebody speaks to them, and who are thrown into an all-consuming panic by a trigger that would seem inconsequential to anybody else. I’ve heard stories that make my life seem like a cakewalk. So when I start feeling sorry for myself after I’ve sat in front of a computer all day making graphs and exploring data, I think:

Andrea, this is your chance to make a difference. This is your chance to help. 

Since beginning work on a project that directly involves patients, my perspective has changed. It has shifted from a focus on achieving great things for myself (i.e. double Nobel laureate) to achieving things that will ultimately help others. If I can have a hand in finding a drug that will help these people get back on their feet and dampen debilitating traumatic memories, I will achieve my own little version of saving the world.


Andrea is a U3 Neuroscience student, and will be posting more in her Breaking Memories series next semester. 

Wanying is a U2 Chemistry student. 

The Chemical Conundrum: How to work efficiently in and out of the lab

By Isabella Liu

Like I said in my first entry, I work in the lab two days a week due to my schedule. Sometimes, when my course-load becomes heavy, I’m unable to come in and run my reactions. To compensate for this, I do a lot of literature searches and ask the necessary questions before coming to the lab to run my reactions; when I do come in during my two days, I make sure that no time is being wasted.

To be honest, after I completed my honours project, there was a period where I felt like I deserved a break. I would go into the lab, re-do some of the reactions I worked on during the summer, and make suChemicalConundrom_Booksre that I got replicable results. People in the lab encouraged that; they all said that I should be thankful that I have this opportunity and enjoy my time in the lab. What they said was partially true, what they didn’t include is the hard work that comes after “enjoying.” October came along and my graduate student was going to graduate. That was when it dawned on me: I was going to have to work independently, solve questions by myself (or ask my Professor Li), all without the help of my graduate student. Her departure gave me a little nudge – it was time for me to start doing my research.

On the topic of literature searches, people often fall into the trap of getting distracted – you’re on your computer, so you automatically get the feeling that whatever you do on your computer is personal. That’s when the personal and work space blur. To counteract this, I make sure that my workspace is as foreign to me as possible (i.e. I don’t put up photographs or have snacks laying around). Basically, I’m invisible.  My desk is also clean – the mess I make on other days is swept away when I write my interim report.

Once your workspace is prepared, finding and organizing relevant articles efficiently is the next challenge. There are several sites that I go for literature searches:

  • Google Scholar (GS) – it’s awesome for recent articles, but if you want to look for specific articles, such as reviews, go to
  • SciFinder (SF) – this was introduced to me by one of my lab courses. You can filter your searches according by authors, the number of cited articles, etc… The only downside is that it is extremely slow, and logs you out quite frequently.  Moreover, it opens up a new window every time you click on the article link.

To balance the advantages and disadvantages of each system, I use GS in conjunction with SF. Once SF gives me the article titles I’m interested in, I go to GS and type in the title. In a split-second (literally, 0.01 second), the article is found and its citation exported.

For exporting citations, GS is awesome. Log into your Google account and head to GS. Click on “settings” in the right-hand corner.

Scroll down to the lowest option – you can choose to export your citations directly according to the type of citation software that you use.

I use BibTex, so every article I search in GS will give me the option of “Import to BibTex.” All I have to do then is copy/paste the BibTex information in a text document. Give it a try – it will change your world!


Look out for more Chemical Conundrum posts from Isabella in January! 


(Photo : By Tom Morris (Own work) [CC-BY-SA-3.0 ( or GFDL (], via Wikimedia Commons)