Understanding Crystallography - Part 2: From Crystals to Diamond 

It´s called Biophysics and you can study that.

lol!

Resonating with that comment. Showing young people science videos like this might be a means of reinvigorating STEM fields.

Bryan, that is a very deep.

do u have any idea how they analyse the image of x-ray to a structure!!

TY for teaching us. And ty for sharing your knowledge.

***** and?

***** so I followed your instructions and now I like country music.

But how do they know the specific atoms based only on the diffraction patterns?

@@lemueljohnurbano4448 I got it from ChatGPT Determining the specific atoms within a crystal structure based solely on diffraction patterns is a complex process that involves several steps and relies on fundamental principles of crystallography, chemistry, and physics. Here's a simplified explanation: 1. **Phase Problem Resolution:** One of the most critical challenges in X-ray crystallography is solving the phase problem. The diffraction pattern only provides information about the amplitude (intensity) of the scattered X-rays, not their phase (relative position). To determine the positions of atoms accurately, scientists must find a way to recover the phase information from the diffraction data. 2. **Fourier Transform:** The diffraction pattern is essentially a Fourier transform of the electron density distribution within the crystal. By applying mathematical techniques, such as Fourier analysis, scientists can convert the diffraction data into an electron density map. 3. **Model Building:** Based on the electron density map, scientists can begin to build a molecular model by placing atoms in the positions that best fit the observed electron density. This process involves trial and error, as well as refinement techniques to improve the fit between the model and the experimental data. 4. **Iterative Refinement:** The initial molecular model is refined iteratively through computational methods to optimize the agreement between the model and the experimental diffraction data. This refinement process adjusts the positions, orientations, and thermal parameters (e.g., atomic displacement parameters) of atoms to minimize discrepancies. 5. **Chemical Constraints:** Chemical knowledge and constraints also play a crucial role in determining the identity and positions of atoms within the crystal structure. For example, the known chemical composition of the sample and the expected bonding patterns based on the molecular formula guide the interpretation of the electron density map and the placement of atoms. 6. **Validation:** The final refined model is subjected to rigorous validation procedures to ensure its accuracy and reliability. This includes checking for geometric consistency, chemical plausibility, and agreement with independent experimental data. Overall, solving the phase problem and determining the specific atoms within a crystal structure based on diffraction patterns involve a combination of experimental data analysis, computational modeling, and chemical reasoning. It requires expertise in crystallography, mathematics, and chemistry to interpret the diffraction data accurately and derive meaningful structural information about the sample.

I love it

I'm not a crystallographer but from what I have come to understand after watching lots of lectures on this technology is that even large molecules will tend to naturally crystallize quite easily as long as a few conditions can be met. One thing that greatly helps the process is that the solution of proteins is as homogenous as possible, but apparently that isn't absolutely necessary. The main catalyst for crystallization is due to additives and precipitants (hope I spelled that right). However, I don't know enough about chemistry to know why those chemicals have such a properties. Anyways, I think I got that info from this same channel, different video.

fullyawakened Getting the ribosome to crystallise was far from easy but it had to be done in order to solve the structure by this approach. It took years of work in many different labs around the world, first to crack bacterial ribosomes which are smaller. The one shown in the film is from yeast, also a single-celled organism but nevertheless a member of the 'eukaryotes', the branch of the tree of life that contains all multi celled organisms, such as ourselves.

Stephen Curry Thanks for the videos, and clarification. This stuff is so incredibly fascinating! These videos have spurred me to look up all kinds of documentaries on crystallography in the last week or so.

fullyawakened Really nice to hear that! I'm sure you'll have seen the RI's Crystallography Collection: www.richannel.org/celebrating-crystallography

@@scurryww The link you provided is now hosting a Trojan according to Malwarebytes.

Lan Fang nothing ingenious about it.

your mom

Nevermind. They do seem to have a 3d software that reconstructs it

The computer doesn't tell you this. In most cases we know it in advance because the amino acid sequence is known and chemists long ago worked out the atomic structure (and chemical composition) of those. The tricky ones to figure out are where you have metal ions bound to your protein; in these cases one can sometimes use X-ray fluorescence to identify the chemical species.

Stephen Curry Thanks for the answer, I thought computer does it all. Well that explains. Out of curiosity, is NMR spectroscopy not suitable enough for structure determination of these biological macromolecules in solution? I think I've attended a lecture about that previous semester, but I didn't pay too much attention, I just thought it must be very complicated. This sure looks more convenient.

Ace0077 NMR is a very useful technique for structure determination, especially if you can't grow crystals (which is a common problem). It gives you the 3D structure but not in quite so much detail as crystallography; plus, it's pretty hard to use NMR to solve the structures of large proteins. On the positive side, it can give information on protein dynamics that you don't get with crystallography. In practice, many structural biology labs use both approaches.

I think that for this technique to work, the X-ray beam needs to be collimated. The collimator is a tube with lots of microtubes in it. The majority of the X-ray photons gets absorbed. Only the straightest path photons make it out of the tube, therefore you lose a lot of brightness. In the case of gamma rays, first you need some methods of producing them. The second problem is that gamma rays are hard to collimate since they are hard to absorb. The collimator would have to be very long.

louis tournas Thank you for the information, this will keep me busy for a while. :)

I could be wrong but I think the longer the accelerator, the more energy have the x rays produced, and that means they have a shorter wavelength than x rays produced in smaller facilites. So in a way they are increasingly going into the path of producing gamma rays, but its use would be more to determine the internal structure of atoms since for molecules x rays this powerful seem enough.

It's 'Porto' by LASERS - you can find it here: freemusicarchive.org/music/lasers/

Thanks

Outro: Ghosts - Grief and Sleep

In this case it's largely because I (who co-wrote the script & presented) am a structural biologist. I'm sure there are some great stories to tell about the physics, chemistry, materials science etc. that goes on at Diamond but I'll leave it to other experts to take up the challenge.

Stephen Curry Now I feel sheepish for venting my (minor) frustrations online! Nonetheless, I thought it was well-written and presented. Disseminating science at this level to the broader community will always be a challenge, and this (along with the recent IUCr videos) are the best I've seen yet. Plus, it's interesting to see how the other half of the synchrotron lives :)

Hugh Simons No worries! You are right in any case that there should be more effort made to cover the physics and chemistry research that comes from synchrotrons (and other particle accelerators). As someone who originally trained in physics, I'm desperate to see Particle Fever: particlefever.com

@@scurryww what you up to nowadays

Academic English :P

Ha ha - well spotted. It's true there are no hydrogen atoms shown in the molecular model that appears at about 6:21. Crystallographers often leave them out because they have many fewer electrons that the main constituents of proteins (C, N, O) and so contribute only weakly to X-ray scattering. They do appear and are included in models constructed using very high resolution data. For example, you can see them emerging as 'bumps' in the photo in this only blogpost which is of an electron density map of the highest resolution structure ever produced in my lab (about 1.45 Å) - occamstypewriter.org/scurry/2008/12/10/come_and_see/ In contrast, protein structures solved by NMR always include H atoms since these contribute very substantially to the data that are recorded in such experiments.

Thank you for the link, interesting data. Actually it was not the model that draw my attention but only the inaccurate statement itself. These details are just something one picks up studying Chemistry and I would be lying if I said I don't enjoy pointing out such inaccuracies ;)

minj4ever Well, I guess what my slip shows is that crystallographers tend to dismiss hydrogens because it is relatively rare for us to have to deal with them. Although the resolution of our data means they are usually invisible, we know they are there and can fairly easily determine their positions from a knowledge of chemistry. However, we're in the habit of not doing that calculation (unless it is necessary for understanding enzyme mechanisms or ligand binding). Personally, I prefer to look at protein structures without the H atoms shown because I think the models look too messy with them included! Of course, NMR spectroscopists tend to take the opposite view.

What? Are you even serious? I'm not even sure why I'm bothering to reply, but: it's a visualization tool. The proof comes from a concrete, mathematical understanding of how light diffracts. The structure is solved from the diffraction pattern by, basically, a inverse Fourier transform of all the diffraction patterns captured. But, most importantly, where do you feel you've gotten the understanding that would be necessary to actually criticize crystallography's position as science? From this short video?

@@BinaryHistory the Babble