Nikon Ti Eclipse stage issues with HCS

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Nikon Ti Eclipse stage issues with HCS

Luke Noble
Hello, 

I'm having some trouble getting a stage correctly calibrated for use with the HCS generator. New to MM, so my troubleshooting powers are rather weak. 

Pixel calibration seems to work fine - i.e., the expected value is arrived at - but (issue 1) the mapping between stage coordinates in HCS and those from the stage seem to be inverted along X. The stage initiates (origin=0,0) at the plate center when powered on, but manually setting the origin (mmc.setAdapterOriginXY(mmc.getXYStageDevice(), 14380.0, 11240.0) doesn't fix this.  I'm attaching my cfg, and an example of a 4-well position list generated by HCS. After setting the origin as above, the well distances seem to be approximately correct, but in reality A2 corresponds to A1 in the position list (A1=A2, B2=B1, B1=B2). And (issue 2), the tile coordinates are basically on top of one another (I'm using the defaults for a 96-well plate at 10x (+1.5x mag) of 7x6 images with 1 micron spacing.

Perhaps unrelated, but before running the above command for the first time, I was getting errors from the HCS generator at every step (no label named ""),  which suggests perhaps an error reading the calibration value? But I haven't seen this since.

Thanks for any advice.

Luke




------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

MMConfig_nikon_LMN.cfg (6K) Download Attachment
PositionList.pos (82K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Nikon Ti Eclipse stage issues with HCS

Nico Stuurman-2
Hi Luke,

> I'm having some trouble getting a stage correctly calibrated for use
> with the HCS generator. New to MM, so my troubleshooting powers are
> rather weak.
>
> Pixel calibration seems to work fine - i.e., the expected value is
> arrived at - but (issue 1) the mapping between stage coordinates in
> HCS and those from the stage seem to be inverted along X.

Micro-Manager relies on the stage adapter to move the stage in the
correct direction (see:
https://micro-manager.org/wiki/Coordinates_and_Directionality).  I don't
know what stage you have, but look at the Device/Property Browser for
stage properties that can change the direction in X.

> The stage initiates (origin=0,0) at the plate center when powered on,
> but manually setting the origin
> (mmc.setAdapterOriginXY(mmc.getXYStageDevice(), 14380.0, 11240.0)
> doesn't fix this.
? Not sure what you mean.  If I remember correctly, the HCS plugin does
not set the origin, you just have to tell it where well A1 (or whichever
well of your choosing in come versions) is.

> I'm attaching my cfg, and an example of a 4-well position list
> generated by HCS. After setting the origin as above, the well
> distances seem to be approximately correct, but in reality A2
> corresponds to A1 in the position list (A1=A2, B2=B1, B1=B2). And
> (issue 2), the tile coordinates are basically on top of one another
> (I'm using the defaults for a 96-well plate at 10x (+1.5x mag) of 7x6
> images with 1 micron spacing.

Try setting the "Spacing Rule" to "Field of View"/
>
> Perhaps unrelated, but before running the above command for the first
> time, I was getting errors from the HCS generator at every step (no
> label named ""),  which suggests perhaps an error reading the
> calibration value? But I haven't seen this since.

Quite possible.  The HCS plugin needs the Core-XYStage and Core-Focus to
be defined, or it may get quite confused.

Best,

Nico


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Nikon Ti Eclipse stage issues with HCS

luke_noble
Hi Nico, 

Thanks, all of that works. And sorry, seems mostly to have been covered by the mailing list/docs, I had thought there was something more basic playing up. 

There now seems (maybe) to be a conflict between orientation used by the HCS and by mM. I can now generate sensible position grids with HCS (with X inverted), but mM seems to want Y inverted (and not X) ... running mmc.setRelativeXYPosition(mmc.getXYStageDevice(), 10000, 10000) from the plate center moves toward H12 rather than left/up as suggested (though if A1 contains 0,0 and we're in the middle then shouldn't -n,-n move up/left toward A1?).

Lastly, I hope, I'm wanting to get timelapse data for a full plate, 2s intervals for 2 minutes, for 4 fields within each well, i.e., the optimal order would be well/pos/time, rather than pos/time to make use of the wait periods. Can't see this handled by MDA, I guess it's something we need to code ourselves? 

Best,
Luke



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Nikon Ti Eclipse stage issues with HCS

Nico Stuurman-2
Hi Luke,

On 3/9/18 10:11 AM, Luke Noble wrote:
> There now seems (maybe) to be a conflict between orientation used by
> the HCS and by mM. I can now generate sensible position grids with HCS
> (with X inverted), but mM seems to want Y inverted (and not X) ...
> running mmc.setRelativeXYPosition(mmc.getXYStageDevice(), 10000,
> 10000) from the plate center moves toward H12 rather than left/up as
> suggested (though if A1 contains 0,0 and we're in the middle then
> shouldn't -n,-n move up/left toward A1?).

The documentation
(https://micro-manager.org/wiki/Coordinates_and_Directionality) mentions
that stage, not the plate.  When you issue the command:
mmc.setRelativeXYPosition(mmc.getXYStageDevice(), 10000, 10000), you
will now notice that the stage moves to the left and top, and as a
consequence, well H12 will end up closer to the objective (rather than
father away).  It is all a matter of perspective;)

So, all is good, and all behaves as intended.

> Lastly, I hope, I'm wanting to get timelapse data for a full plate, 2s
> intervals for 2 minutes, for 4 fields within each well, i.e., the
> optimal order would be well/pos/time, rather than pos/time to make use
> of the wait periods. Can't see this handled by MDA, I guess it's
> something we need to code ourselves?

I think that you are correct.  Regretfully, adding this option involves
modifying the acquisition engine, which is written in Clojure, so that
there are only very few people who can touch that code.  This should be
easy to handle in a beanshell script.  I'll make this an example
exercise for the Focus On microscopy workshop that I'll be giving in 2
weeks;)

Best,

Nico


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Nikon Ti Eclipse stage issues with HCS

luke_noble
Hi Niko, 

Ha, specimen-centric biologist here. 

OK, this should be fun. I'm confident we can get something workable in less than two weeks (but if I'm overconfident I'll be in touch in 15 days ;-).

Thanks again for the great software and support, it's been a remarkably easy transition for something so complex.

Luke

 

Luke Noble

Postdoctoral Associate
Rockman laboratory
Center for Genomics and Systems Biology
Department of Biology
New York University


On 9 March 2018 at 14:29, Nico Stuurman <[hidden email]> wrote:
Hi Luke,

On 3/9/18 10:11 AM, Luke Noble wrote:
There now seems (maybe) to be a conflict between orientation used by the HCS and by mM. I can now generate sensible position grids with HCS (with X inverted), but mM seems to want Y inverted (and not X) ... running mmc.setRelativeXYPosition(mmc.getXYStageDevice(), 10000, 10000) from the plate center moves toward H12 rather than left/up as suggested (though if A1 contains 0,0 and we're in the middle then shouldn't -n,-n move up/left toward A1?).

The documentation (https://micro-manager.org/wiki/Coordinates_and_Directionality) mentions that stage, not the plate.  When you issue the command: mmc.setRelativeXYPosition(mmc.getXYStageDevice(), 10000, 10000), you will now notice that the stage moves to the left and top, and as a consequence, well H12 will end up closer to the objective (rather than father away).  It is all a matter of perspective;)

So, all is good, and all behaves as intended.

Lastly, I hope, I'm wanting to get timelapse data for a full plate, 2s intervals for 2 minutes, for 4 fields within each well, i.e., the optimal order would be well/pos/time, rather than pos/time to make use of the wait periods. Can't see this handled by MDA, I guess it's something we need to code ourselves?

I think that you are correct.  Regretfully, adding this option involves modifying the acquisition engine, which is written in Clojure, so that there are only very few people who can touch that code.  This should be easy to handle in a beanshell script.  I'll make this an example exercise for the Focus On microscopy workshop that I'll be giving in 2 weeks;)


Best,

Nico


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Question regarding language settings on MAC

Christian Bayer
In reply to this post by Nico Stuurman-2
Hi Nico,

we have managed to build MM 1.4 on Mac OS High Sierra. Thanks a lot for the good and detailed instructions!

During the process of configuration using the hardware config wizard we encounter the problem that floating point numbers are displayed using a „,“ (comma) because we run a German OS. These number format is not accepted by the config wizard. Do you have any hints to overcome this problem?

Thanks a lot and best regards,
Christian
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general