Electronics-Related.com
Forums

Kinesin Walking

Started by John Larkin May 26, 2023
https://www.youtube.com/watch?v=YAva4g3Pk6k

https://www.youtube.com/watch?v=WFCvkkDSfIU

Walking proteins are at about 8:00. Very cool.

All these bits are way smaller than light can resolve, so the people
doing this need signal processing, and they are not EEs.

On Saturday, May 27, 2023 at 12:51:24 AM UTC+10, John Larkin wrote:
> https://www.youtube.com/watch?v=YAva4g3Pk6k > > https://www.youtube.com/watch?v=WFCvkkDSfIU > > Walking proteins are at about 8:00. Very cool. > > All these bits are way smaller than light can resolve, so the people doing this need signal processing, and they are not EEs.
It's a twelve year old exercise in molecular modelling. The molecular structures were originally sorted out by X-ray diffaction. X-rays are a form of electromagnetic radiation, and can be a seen as a sort of very short-wavelength light which actually can resolve the separate atoms involved. The people doing the work are molecular biologists, and not EE's which is the one point that John Larkin has got right. The signal processing - such as it is - happens on big computers, and the process of sequencing proteins is fairly computer intensive, but calling it signal processing is a long stretch. There aren't any finite impulse response filters involved. When I was first working in the UK one of my acquaintances from Melbourne was doing a post-doc at Max Perutz's Laboratory of Molecular Biology in Cambridge where he'd worked out the structure of haemoglobin a few years earlier. She's now a professor of Molecular Biology back in Melbourne and got some kind of gong (government awarded honour) last year. The clip predates Google's solving the protein folding problem. This article is from 2022. That was a much cooler trick. https://www.analyticsinsight.net/deepminds-alphafold2-solves-the-long-stood-protein-folding-problem/ -- Bill Sloman, Sydney
On Friday, May 26, 2023 at 10:51:24 AM UTC-4, John Larkin wrote:
> https://www.youtube.com/watch?v=YAva4g3Pk6k > > https://www.youtube.com/watch?v=WFCvkkDSfIU > > Walking proteins are at about 8:00. Very cool. > > All these bits are way smaller than light can resolve, so the people > doing this need signal processing, and they are not EEs.
Fluorescence microscopy has been in the works for a few decades now. They tag the target with fluorescent molecule protein bond of some kind, and that gets illuminated. The MINFLUX microscope which the researchers now introduce can record protein movements with a spatiotemporal precision of up to 1.7 nanometers per millisecond. It requires the detection of only about 20 photons emitted by the fluorescent molecule. https://phys.org/news/2023-03-major-advance-super-resolution-fluorescence-microscopy.html
On Fri, 26 May 2023 09:21:44 -0700 (PDT), Fred Bloggs
<bloggs.fredbloggs.fred@gmail.com> wrote:

>On Friday, May 26, 2023 at 10:51:24?AM UTC-4, John Larkin wrote: >> https://www.youtube.com/watch?v=YAva4g3Pk6k >> >> https://www.youtube.com/watch?v=WFCvkkDSfIU >> >> Walking proteins are at about 8:00. Very cool. >> >> All these bits are way smaller than light can resolve, so the people >> doing this need signal processing, and they are not EEs. > >Fluorescence microscopy has been in the works for a few decades now. They tag the target with fluorescent molecule protein bond of some kind, and that gets illuminated.
Yes. One can resolve the location of the centroid of a glowing blob of light to a tiny fraction of the wavelength of light. It's a cool trick.
> >The MINFLUX microscope which the researchers now introduce can record protein movements with a spatiotemporal precision of up to 1.7 nanometers per millisecond. It requires the detection of only about 20 photons emitted by the fluorescent molecule. > >https://phys.org/news/2023-03-major-advance-super-resolution-fluorescence-microscopy.html
Needs electronics!
On Saturday, May 27, 2023 at 2:21:49&#8239;AM UTC+10, Fred Bloggs wrote:
> On Friday, May 26, 2023 at 10:51:24&#8239;AM UTC-4, John Larkin wrote: > > https://www.youtube.com/watch?v=YAva4g3Pk6k > > > > https://www.youtube.com/watch?v=WFCvkkDSfIU > > > > Walking proteins are at about 8:00. Very cool. > > > > All these bits are way smaller than light can resolve, so the people > > doing this need signal processing, and they are not EEs. > > Fluorescence microscopy has been in the works for a few decades now. They tag the target with fluorescent molecule protein bond of some kind, and that gets illuminated. > > The MINFLUX microscope which the researchers now introduce can record protein movements with a spatiotemporal precision of up to 1.7 nanometers per millisecond. It requires the detection of only about 20 photons emitted by the fluorescent molecule. > > https://phys.org/news/2023-03-major-advance-super-resolution-fluorescence-microscopy.html
Fluorescence microscopy is just conventional microscopy. Confocal microscopes do have advantages but nothing dramatic https://pubmed.ncbi.nlm.nih.gov/20862051/ describes a rather more practical confocal microscope than the original devices. One of the authors - Martin Harris - is an old friend of mine and made quite a lot of money out of it. There are some weird trick you can play - like two-photon excitation - which can push up the resolution appreciably. Stefan Hell takes this even further. "The basic principles and standard implementations of STED (stimulated emission depletion) and RESOLFT (reversible saturable/switchable optical linear (fluorescence) transitions) microscopy" are very weird indeed. Since this all post-dates the youtube presentation that John Larkin posted a link to, it's not relevant to it, though it does seem to be able to validate the original computer modelling. To some extent the development of this kind of hardware is driven by the ambition to see what we had already worked out had to be happening at that scale. -- Bil Sloman, Sydney
On Friday, May 26, 2023 at 1:34:04&#8239;PM UTC-4, Anthony William Sloman wrote:
> On Saturday, May 27, 2023 at 2:21:49&#8239;AM UTC+10, Fred Bloggs wrote: > > On Friday, May 26, 2023 at 10:51:24&#8239;AM UTC-4, John Larkin wrote: > > > https://www.youtube.com/watch?v=YAva4g3Pk6k > > > > > > https://www.youtube.com/watch?v=WFCvkkDSfIU > > > > > > Walking proteins are at about 8:00. Very cool. > > > > > > All these bits are way smaller than light can resolve, so the people > > > doing this need signal processing, and they are not EEs. > > > > Fluorescence microscopy has been in the works for a few decades now. They tag the target with fluorescent molecule protein bond of some kind, and that gets illuminated. > > > > The MINFLUX microscope which the researchers now introduce can record protein movements with a spatiotemporal precision of up to 1.7 nanometers per millisecond. It requires the detection of only about 20 photons emitted by the fluorescent molecule. > > > > https://phys.org/news/2023-03-major-advance-super-resolution-fluorescence-microscopy.html > Fluorescence microscopy is just conventional microscopy. Confocal microscopes do have advantages but nothing dramatic > > https://pubmed.ncbi.nlm.nih.gov/20862051/ > > describes a rather more practical confocal microscope than the original devices. One of the authors - Martin Harris - is an old friend of mine and made quite a lot of money out of it. > > There are some weird trick you can play - like two-photon excitation - which can push up the resolution appreciably. > > Stefan Hell takes this even further. "The basic principles and standard implementations of STED (stimulated emission depletion) and RESOLFT (reversible saturable/switchable optical linear (fluorescence) transitions) microscopy" are very weird indeed. > > Since this all post-dates the youtube presentation that John Larkin posted a link to, it's not relevant to it, though it does seem to be able to validate the original computer modelling. To some extent the development of this kind of hardware is driven by the ambition to see what we had already worked out had to be happening at that scale.
It seems like every day there's a new discovery with all of this high precision instrumentation, once they figure out what they're looking at. Prior work, theorizing, speculation, etc, prepares the researchers to make these discoveries. Stating the obvious here of course. Chance favors only the prepared mind. - Louis Pasteur.
> > -- > Bil Sloman, Sydney
On Saturday, May 27, 2023 at 3:03:27&#8239;AM UTC+10, John Larkin wrote:
> On Fri, 26 May 2023 09:21:44 -0700 (PDT), Fred Bloggs > <bloggs.fred...@gmail.com> wrote: > >On Friday, May 26, 2023 at 10:51:24?AM UTC-4, John Larkin wrote: > >> https://www.youtube.com/watch?v=YAva4g3Pk6k > >> > >> https://www.youtube.com/watch?v=WFCvkkDSfIU > >> > >> Walking proteins are at about 8:00. Very cool. > >> > >> All these bits are way smaller than light can resolve, so the people > >> doing this need signal processing, and they are not EEs. > > > >Fluorescence microscopy has been in the works for a few decades now. They tag the target with fluorescent molecule protein bond of some kind, and that gets illuminated. > > Yes. One can resolve the location of the centroid of a glowing blob of light to a tiny fraction of the wavelength of light. It's a cool trick.
That's probably not a good description of what's going on. It's more setting up the excitation so that fluorescence can only be stimulated inside a very small volume. The molecules involved have a nasty habit of coming to pieces if they see too many excitation photons. Single molecule fluoresence microscopy tends to kill off the molecule after a few excitations.
> >The MINFLUX microscope which the researchers now introduce can record protein movements with a spatiotemporal precision of up to 1.7 nanometers per millisecond. It requires the detection of only about 20 photons emitted by the fluorescent molecule. > > > >https://phys.org/news/2023-03-major-advance-super-resolution-fluorescence-microscopy.html > > Needs electronics!
It certainly uses quite a lot of electronics, some of it developed by physicists. -- Bill Sloman, Sydney
On Saturday, May 27, 2023 at 3:47:14&#8239;AM UTC+10, Fred Bloggs wrote:
> On Friday, May 26, 2023 at 1:34:04&#8239;PM UTC-4, Anthony William Sloman wrote: > > On Saturday, May 27, 2023 at 2:21:49&#8239;AM UTC+10, Fred Bloggs wrote: > > > On Friday, May 26, 2023 at 10:51:24&#8239;AM UTC-4, John Larkin wrote:
<snip>
> > There are some weird trick you can play - like two-photon excitation - which can push up the resolution appreciably. > > > > Stefan Hell takes this even further. "The basic principles and standard implementations of STED (stimulated emission depletion) and RESOLFT (reversible saturable/switchable optical linear (fluorescence) transitions) microscopy" are very weird indeed. > > > > Since this all post-dates the youtube presentation that John Larkin posted a link to, it's not relevant to it, though it does seem to be able to validate the original computer modelling. To some extent the development of this kind of hardware is driven by the ambition to see what we had already worked out had to be happening at that scale. > > It seems like every day there's a new discovery with all of this high precision instrumentation, once they figure out what they're looking at. Prior work, theorizing, speculation, etc, prepares the researchers to make these discoveries. Stating the obvious here of course. > > Chance favors only the prepared mind. - Louis Pasteur.
There's no chance element in refining an instrument to let you see what you expect to see. It gets interesting when it lets you see something unexpected, but that doesn't seem to have happened yet. It would be worth posting here when it did. -- Bill Sloman, Sydney