What is typically reported as “the” resolution is really just the *overall* resolution. But it’s really important to appreciate that local resolution may be worse (sometimes way worse!) than the overall resolution that is reported. Imagine taking a picture of a track meet. Even if you had a great camera, the runner might look blurry even though the spectators look crisp. Similarly, although the physics behind it are different, regions of a map corresponding to parts of the molecule that are more flexible or dynamic (often loops and things on the surface) might look more sausagey. And thus harder (or even impossible) to build a model into. So often those atoms are left unmodeled and you can see which are missing if you look at the PDB file (more on this later). Sometimes scientists will show this as dashed lines in their models.
For example, sometimes whole amino acids in proteins (especially the termini (chain ends) are left unmodeled. And sometimes you can just make out the backbone (less flexible) but can’t make out the side chain so it’s only partially modeled or an alanine is modeled on instead of the amino acid that’s actually there since alanine is pretty “generic” structure-wise. More on that in my alanine post. bit.ly/alaninechirality
Bottom line thing to keep in mind is that scientists can make out details better in some areas of the map than in others, and thus can model those regions more confidently. These regions have a low B-factor (aka temperature or displacement factor). On the other hand, regions with a high B-factor have more "iffy" data - it's less clear how to model in the underlying atoms. If you view a structure in something like PyMol, Chimera, or even the PDB web browser now, you can color the model by B-factor. Typically it’s displayed as a heat map with red corresponding to high B factor. You tend to see a lot of red on the periphery of the protein a blue inside.
No matter where in a molecule some feature is located, even if it has a low B factor, you want to make sure you actually view the maps yourself before relying on the model if you are trying to interpret some claim about it, or make some hypothesis or something. So I made a video explaining how to do this. Hope it helps.
Before you judge a structure based on its resolution, know that resolution isn’t the only thing that matters and even a low resolution structure can provide valuable information, especially if used in concert with high res info. Such as, for example, placing models from high res crystal structures of individual complex components into a low res cryo map of the whole complex.
As for the biophysics going on behind the scene leading to the resolution that is seen? Well, it’s different for x-ray crystallography and cryo-EM. And it’s also different how they determine it. I’m more familiar with how it’s done in crystallography so I will describe in detail below (adapted from older post). But if you want to learn more about the cryo-EM stuff, I talk about it some in the video and here are some links to papers I recommend.
Beckers, M., Mann, D., & Sachse, C. (2021). Structural interpretation of cryo-EM image reconstructions. Progress in biophysics and molecular biology, 160, 26-36. doi.org/10.1016/j.pbiomolbio.2020.07.004
Marques, M. A., Purdy, M. D., & Yeager, M. (2019). CryoEM maps are full of potential. Current opinion in structural biology, 58, 214-223. doi.org/10.1016/j.sbi.2019.04.006
D'Imprima, E., & Kühlbrandt, W. (2021). Current limitations to high-resolution structure determination by single-particle cryoEM. Quarterly reviews of biophysics, 54, e4. doi.org/10.1017/S0033583521000020 more on understanding crystal structures: bit.ly/crystalstructuremodels & th-cam.com/video/YK3VkqD2o2s/w-d-xo.html
more on cryo-EM here: bit.ly/cryoemxray & bit.ly/cryoEMbumblyintro & th-cam.com/video/qcS6oZlEtH0/w-d-xo.html
Pédelacq, J. D., Cabantous, S., Tran, T., Terwilliger, T. C., & Waldo, G. S. (2006). Engineering and characterization of a superfolder green fluorescent protein. Nature biotechnology, 24(1), 79-88. doi.org/10.1038/nbt1172
pdb101.rcsb.org/motm/42
more about all sorts of things: #365DaysOfScience All (with topics listed) 👉 bit.ly/2OllAB0 or search blog: thebumblingbiochemist.com
What is typically reported as “the” resolution is really just the *overall* resolution. But it’s really important to appreciate that local resolution may be worse (sometimes way worse!) than the overall resolution that is reported. Imagine taking a picture of a track meet. Even if you had a great camera, the runner might look blurry even though the spectators look crisp. Similarly, although the physics behind it are different, regions of a map corresponding to parts of the molecule that are more flexible or dynamic (often loops and things on the surface) might look more sausagey. And thus harder (or even impossible) to build a model into. So often those atoms are left unmodeled and you can see which are missing if you look at the PDB file (more on this later). Sometimes scientists will show this as dashed lines in their models.
For example, sometimes whole amino acids in proteins (especially the termini (chain ends) are left unmodeled. And sometimes you can just make out the backbone (less flexible) but can’t make out the side chain so it’s only partially modeled or an alanine is modeled on instead of the amino acid that’s actually there since alanine is pretty “generic” structure-wise. More on that in my alanine post. bit.ly/alaninechirality
Bottom line thing to keep in mind is that scientists can make out details better in some areas of the map than in others, and thus can model those regions more confidently. These regions have a low B-factor (aka temperature or displacement factor). On the other hand, regions with a high B-factor have more "iffy" data - it's less clear how to model in the underlying atoms. If you view a structure in something like PyMol, Chimera, or even the PDB web browser now, you can color the model by B-factor. Typically it’s displayed as a heat map with red corresponding to high B factor. You tend to see a lot of red on the periphery of the protein a blue inside.
No matter where in a molecule some feature is located, even if it has a low B factor, you want to make sure you actually view the maps yourself before relying on the model if you are trying to interpret some claim about it, or make some hypothesis or something. So I made a video explaining how to do this. Hope it helps.
Before you judge a structure based on its resolution, know that resolution isn’t the only thing that matters and even a low resolution structure can provide valuable information, especially if used in concert with high res info. Such as, for example, placing models from high res crystal structures of individual complex components into a low res cryo map of the whole complex.
As for the biophysics going on behind the scene leading to the resolution that is seen? Well, it’s different for x-ray crystallography and cryo-EM. And it’s also different how they determine it. I’m more familiar with how it’s done in crystallography so I will describe in detail below (adapted from older post). But if you want to learn more about the cryo-EM stuff, I talk about it some in the video and here are some links to papers I recommend.
Beckers, M., Mann, D., & Sachse, C. (2021). Structural interpretation of cryo-EM image reconstructions. Progress in biophysics and molecular biology, 160, 26-36. doi.org/10.1016/j.pbiomolbio.2020.07.004
Marques, M. A., Purdy, M. D., & Yeager, M. (2019). CryoEM maps are full of potential. Current opinion in structural biology, 58, 214-223. doi.org/10.1016/j.sbi.2019.04.006
D'Imprima, E., & Kühlbrandt, W. (2021). Current limitations to high-resolution structure determination by single-particle cryoEM. Quarterly reviews of biophysics, 54, e4. doi.org/10.1017/S0033583521000020
more on understanding crystal structures: bit.ly/crystalstructuremodels & th-cam.com/video/YK3VkqD2o2s/w-d-xo.html
more on cryo-EM here: bit.ly/cryoemxray & bit.ly/cryoEMbumblyintro & th-cam.com/video/qcS6oZlEtH0/w-d-xo.html
intro to PDB, crystal structure entries, crystal contents, resolution: bit.ly/pdbstructures & th-cam.com/video/Re2gwi-_OEw/w-d-xo.html & th-cam.com/video/IZtHsUFbyes/w-d-xo.html
links to more about the GFP example structure:
www.rcsb.org/structure/2b3p
Pédelacq, J. D., Cabantous, S., Tran, T., Terwilliger, T. C., & Waldo, G. S. (2006). Engineering and characterization of a superfolder green fluorescent protein. Nature biotechnology, 24(1), 79-88. doi.org/10.1038/nbt1172
pdb101.rcsb.org/motm/42
more about all sorts of things: #365DaysOfScience All (with topics listed) 👉 bit.ly/2OllAB0 or search blog: thebumblingbiochemist.com
Thanks a lot for the video. Wish you a very healthy and successful year ahead. Happy New Year.
Thank you! Glad you found it helpful! Wishing you a happy and healthy new year as well
thank you so so much for this video! immensly useful when preparing for my XRD exam
Glad it was helpful!