3 ms·

My understanding as an x-ray crystallographer and electron microscopist is that XFEL has been nothing short of a massive disappointment. Yes, it works in theory

My understanding as an x-ray crystallographer and electron microscopist is that XFEL has been nothing short of a massive disappointment. Yes, it works in theory, but in practice it's entirely inefficient and doesn't address the current bottlenecks of structural biology at all. I'm not saying this to downplay the project (I know very little about it) but the reception among those in the field it's supposedly applicable for has been underwhelming.

djaque 6y ago
I'm not really sure about crystallography (not my field), but I think it's maybe an issue with expectation management. Everyone is trying to sell the new technology as fix for every field. In reality XFEL experiments do have a lot of overhead associated with them and stuff like synchrotron light sources are probably better suited for "everyday" type measurements. The real breakthroughs that I was talking about are for time resolved and "diffract before destroy" experiments for delicate samples. Those ones can't really be done at a synchrotron with long bunches. I'm especially excited for femtosecond time resolved measurements of chemical reactions. It's going to be so cool to watch chemical reactions occur at an atomic scale. I guess the other cool application is that there is so much light that you can perform diffraction off of single molecules when you want your sample in solution/gas phase or can't form crystals. Correct me if I'm wrong, but my impression was that the intensity of XFELs also enabled that.
- COGlory 6y ago
  I didn't know about the time resolved stuff, and thats the type of thing that big expensive projects seem much more economical for. Presumably we will learn foundational (previously theoretical) information which will have wide, lasting effects. The project is so staggeringly complicated that when I've talked to people involved, they knew very little outside of their own minor field (for instance, microfluidics to time droplets with pulses). The way that the project is being sold from the structural side is that you won't need a crystal, which is obviously huge. However, there's the issue of scale, and also the classic issue of... We can't refocus diffracted X rays so we're still limited by protein we can express in a selenium doped media, since you need the selenium signal to determine phase data from the diffractions. The list of proteins that are soluble but won't crystallize/can't be resolved by Cryo-EM is pretty small, and so it's an incremental upgrade no one will have access to, at best. At least, from the structural side.
pfdietz 6y ago
I get the impression that cryo electron microscopy is now the big new(ish) technique for structure determination.
- COGlory 6y ago
  Yes, advances in the detectors (essentially CMOS electron detectors) have absolutely revolutionized the field and solved many of the major problems. However, there are still limitations, especially with smaller molecules.

djaque 6y ago

I'm not really sure about crystallography (not my field), but I think it's maybe an issue with expectation management. Everyone is trying to sell the new technology as fix for every field. In reality XFEL experiments do have a lot of overhead associated with them and stuff like synchrotron light sources are probably better suited for "everyday" type measurements. The real breakthroughs that I was talking about are for time resolved and "diffract before destroy" experiments for delicate samples. Those ones can't really be done at a synchrotron with long bunches. I'm especially excited for femtosecond time resolved measurements of chemical reactions. It's going to be so cool to watch chemical reactions occur at an atomic scale. I guess the other cool application is that there is so much light that you can perform diffraction off of single molecules when you want your sample in solution/gas phase or can't form crystals. Correct me if I'm wrong, but my impression was that the intensity of XFELs also enabled that.

COGlory 6y ago
I didn't know about the time resolved stuff, and thats the type of thing that big expensive projects seem much more economical for. Presumably we will learn foundational (previously theoretical) information which will have wide, lasting effects. The project is so staggeringly complicated that when I've talked to people involved, they knew very little outside of their own minor field (for instance, microfluidics to time droplets with pulses). The way that the project is being sold from the structural side is that you won't need a crystal, which is obviously huge. However, there's the issue of scale, and also the classic issue of... We can't refocus diffracted X rays so we're still limited by protein we can express in a selenium doped media, since you need the selenium signal to determine phase data from the diffractions. The list of proteins that are soluble but won't crystallize/can't be resolved by Cryo-EM is pretty small, and so it's an incremental upgrade no one will have access to, at best. At least, from the structural side.

pfdietz 6y ago

I get the impression that cryo electron microscopy is now the big new(ish) technique for structure determination.

COGlory 6y ago
Yes, advances in the detectors (essentially CMOS electron detectors) have absolutely revolutionized the field and solved many of the major problems. However, there are still limitations, especially with smaller molecules.