Pre-MSI Set-up¶

Design Presets¶

To start your own MSI experiment, you must decide what presets you want to use. Presets are used to define scope and camera parameters with which images are acquired.

MSI presets can be customized to your need of data collection. By matching nodes/subnodes in MSI with presets and move types, the behavior of the node is defined. Please do not use "-" in preset name It is reserved for frame aligned sum preset labeling.

For easy reference and for preloading node settings in the our distribution, preset names in MSI has been standardized at NRAMM. We strongly advice that you use these to start with so that the default settings will work right away. You can use other names when you become advanced user and are comfortable changing settings. They are abbreviated to 2 letter codes to reduce the length of the filenames that contains every preset used in its family history.

Example MSI preset nomenclature:

Preset name:	abbrev for:	in the context of:
gr	grid	em grid
sq	square	em grid square
hl	hole	quantifoil hole
fc	focus	focus image to be checked with fft
fa	focus-auto	automatic focusing
en	exposure-near	close-to-zero defocus exposure image in a focal pair
ef	exposure-far	far-from-zero defocus exposure image in a focal pair

Example MSI preset parameters (based on 120 or 200 kV high tension, 100 um C2 aperture and 100 um objective aperture on a 4k ccd with pixel size at 1.6 angstrom at scope nominal magnification of 50,000x):

Magnification:	Preset name:	Image Shift (x,y):	Dimension:	Binning:	Beam Coverage:	Exposure Time (ms):	Spot Size:	Defocus (m):
120	gr	Aligned	512	8	max	20	4	0.0
550	sq	Aligned	1024	4	1x CCD size	100	4	-2e-3
5000	hl	Aligned	512	8	1x CCD	20	4	-1.5e-4
50000	fc	0,0	512	1	<~ 1x CCD	300	4	-2e-6
50000	fa	0,0	1024	4	>2x CCD	50	4	-2e-6
50000	en	0,0	4096	1	2x CCD	170 (10e/A^2)	4	-1e-6
50000	ef	0,0	4096	1	2x CCD	170 (10e/A^2)	4	-2e-6

The preset parameters in this example are chosen to give the following properties:

• en and ef are for final exposures and therefore have highest resolution and dimension.

• en stands for near focus exposure and therefore has lower defocus value than ef,the far-from-focus exposure.

• fa is for high resolution autofocus and therefore need to be high mag. The dimension and binning is a result of compromising speed, S/N ratio and sensitivity to
  details.

• fc is for manual focus and ice melting and therefore more intense than fa. To get good FFT for checking Thon ring behavior, it is not binned. The small dimension is for increasing speed.

• hl is for intermediate targeting. It needs to cover the error range of modeled stage position to allow final targeting by image shift only, hence the magnification. The binning is for speed.

• sq is for grid square targeting. The image produced by the preset ideally need to cover most if not all of the square but with enough resolution to evaluate the content. However, the first priority is to choose a magnification at which no aperture used during the data collection will limit much of the illuminated area on the CCD. If too much of the area is obstructed, target adjustment critical to MSI operation will fail. The coverage of the grid square can always be achieved by choosing multiple targets on it.

• gr is for producing grid atlas. It is selected based on the size number of tiles needed to cover the whole grid in an acceptable time and the amount of distortion common to very low mag imaging. Objective aperture can be removed and a larger C2 aperture used while using this preset to acquire grid atlas. It is never used after the atlas is acquired.

In designing your own preset, follow the above properties using a mag that gives the required coverage with your scope, camera, and grid mesh. If you have a 2k or 1k camera, binning may not be as necessary as for a 4k camera whose data acquisition time is 10-30 sec without binning.

Another Recent Preset Design Example (Rotavirus 4.4 A resolution, Cambell et. al., Structure(2012))¶

Tecnai F20 200 kV high tension, Gun Lens 3 extraction voltage 4300, 50 um C2 aperture and 100 um objective aperture (except when obtaining grid atlas).
Camera is a Tietz F415 4k ccd with pixel size at 1.6 angstrom at scope nominal magnification of 50,000x for lower mag MSI and focusing, and a Direct Electron DE-12 4kx3k direct detection camera for final exposure (ed preset):

Magnification:	Preset name:	Image Shift (x,y):	Dimension:	Binning:	Beam Coverage:	Exposure Time (ms):	Dose (e/A^2):	Spot Size:	Defocus (m):	C2 (um):
120	gr	Aligned	512	8	max	20	n/a	4	0.0	100
1700	sq	Aligned	1024	4	1x CCD size	10	n/a	4	-2e-4	50
5000	hl	Aligned	1024	4	1x CCD	7	0.004	4	-1.5e-4	50
100000	fc	0,0	512	2	<~ 1x CCD	200	164	4	-2.9e-7	50
100000	fa	0,0	1024	4	~4x CCD	50	2.5	4	-1.5e-6	50
100000	en	0,0	4096	1	1.3 um on the specimen	407	20	4	(-1e-6 to -2e-6)	50
29000	ed	0,0	3kx4k	1	1.3 um on the specimen	407	20	4	(-1e-6 to -2e-6)	50

• Note: This experiment used direct detection camera for final exposure (ed preset). en preset was not used in the actual experiment. It is here just for reference and to show what we would have used if recorded on CCD based on our GroEL study. During process, the final images for en would have been binned by 2 based on our experience so the en preset could have being collected at 2048 binned by 2.

• Note: no defocal pair was aquired in this experiment. Therefore, no ef preset.

Preset Design Example for Falcon II on Krios¶

Tecnai Titan Krios 300 kV high tension, Gun Lens 3 extraction voltage 4150, 70 um C2 aperture and 100 um objective aperture (except when obtaining grid atlas).
Falcon II mounted with Gatan Orius as the secondary camera. Pixel size at 1.4 angstrom at scope nominal magnification of 59,000x (ed preset).
The c-flat grid used in this has 1.2 um holes. One image is taken at each hole.

Magnification:	Preset name:	Camera:	Image Shift (x,y):	Dimension:	Binning:	Beam Coverage:	Exposure Time (ms):	Dose (e/A^2):	Spot Size:	Defocus (m):	C2 (um):
40	gr	Orius	Aligned	1024	4	max	100	n/a	10	0.0	70
1700	sq	Falcon	Aligned	1024	4	1x CCD size	100	n/a	10	-5e-4	70
3800	hl	Falcon	Aligned	1024	4	12 um	300	n/a	10	-2e-4	70
45000	fm	Orius	0,0	1024	2	small	500	100	7	-2.9e-7	70
75000	fc	Orius	0,0	1024	2	0.5 um on the specimen	500	164	6	-2.9e-7	70
75000	fa	Falcon	0,0	1024	4	1.2 um on the specimen	200	2.5	7	-1.5e-6	70
59000	en	Falcon	0,0	4kx4k	1	1.3 um on the specimen	2000	45	6	(-1e-6 to -2e-6)	70

• fm is the melting preset. No image is taken with this. The preset is used only to set microscope parameters. The camera is set to Orius as extra protection.
• Frames are saved in ed preset with (1,2,17) set in "Frames to use" field, giving 22.5 e/A^2 total dose in the movie that is used for image processing and volume reconstruction.

Preset Design Example for small pixel Gatan K2 camera alone¶

Krios-K2 Presets used for 2.8 A proteasome reconstruction at TSRI¶

Tecnai Krios 300 kV high tension, Gun Lens 5 extraction voltage 4300, 70 um C2 aperture and 100 um objective aperture (except when obtaining grid atlas).
Gatan camera dimension 3710(w)x 3838(h) after rotation to Leginon standard. The size of the beam on the scope main viewing screen always covers the 1.2 um hole at en preset on our scope.
Pixel size is 1.21 A at 22500x scope nominal mag.:

Magnification:	Preset name:	Image Shift (x,y):	Dimension:	Binning:	Camera Dose (e/camera pixel/s):	Exposure Time (ms):	Specimen Dose Rate (e/A^2):	Spot Size:	Defocus (m):	C2 (um):	frame saving:
81	gr	Aligned	927x959	4	8	100	n/a	10	0.0	70	no
165	sq	Aligned	927x959	4	8	100	n/a	9	-2e-4	70	no
1700	hl	Aligned	927x959	4	8	300	0.008	9	-1.5e-4	70	no
22500	fc	0,0	924x924	2	>=8*	500	8.8	6	-2.9e-7	70	no
22500	fm	0,0	n/a	n/a	high	n/a	high	5	n/a	70	no
22500	fa	0,0	927x959	4	>=8*	200	3.5	6	-1.5e-6	70	no
22500	en	0,0	7420x7676	1	8 e/physical pixel/s and ~1.9 um on the specimen	7600	36	8	(-1e-6 to -2e-6)	70	200 ms/frame

• Note: This experiment has an additional preset fm used for melting the ice exclusively. The screen goes down during the melt, therefore, it does not cause radiation damage on the camera. Do not attempt to acquire an image with this preset.

• Note: If the correlation or power spectrum don't work well enough for the standard Camera Dose Rate of 10 e/pixel/s for fc and fa, you may increase it more so not to have to use longer exposure time. The recommended value in literature is meant to give highest DQE at all sampling frequencies. Using a higher camera dose by 50% will certainly not damage the sensor.

• Note: no defocal pair was acquired in this experiment. Therefore, no ef preset.

J2100f k2 preset design¶

Magnification:	Preset name:	Image Shift (x,y):	Dimension:	Binning:	Camera Dose (e/camera pixel/s):	Exposure Time (ms):	Specimen Dose Rate (e/A^2):	Spot Size:	Defocus (m):	C2 (um):	frame saving:
80	gr	Aligned	927x959	4	10	100	n/a	5	0.0	100	no
800	sq	Aligned	927x959	4	10	100	n/a	3	0.0	50	no
2500	hl	Aligned	927x959	4	10	400	0.008	4	-2e-5	50	no
2500	hc	Aligned	927x959	4	10	400	0.008	4	-2e-5	50	no
25000	fc	0,0	926x926	2	>=10*	400	8.8	4	-2.9e-7	50	no
25000	fm	0,0	n/a	n/a	high	n/a	high	4	n/a	50	no
25000	fa	0,0	927x959	4	>=10*	200	3.5	4	-1.5e-6	50	no
30000	ed	0,0	3710x3838	1	8 e/pixel/s and ~1.3 um on the specimen	5000	27	4	(-1e-6 to -2e-6)	50	200 ms/frame

• Spot 5 on this scope is modified by JEOL engineers to reduce the beam intensity further from the original values so that it does not over dose K2 sensor.
• The strong image shift hysteresis of the 800x square preset required us to use MSI application that has an extra step of centering the holes using 2500x mag. See Leginon_Operation for_JEM_scopes

Preset Design Example for FEI Krios EF-TEM¶

EF-TEM Krios-K2-Ceta Presets¶

Tecnai Krios 300 kV high tension always in EF-TEM mode, Gun Lens 5 extraction voltage 4300, 70 um C2 aperture and 100 um objective aperture (except when obtaining grid atlas).
Gatan camera dimension 3710(w)x 3838(h) after rotation to Leginon standard. The size of the beam on the scope main viewing screen always covers the 1.2 um hole at en preset on our scope.

Pixel size is 1.06 A at 130,000x scope nominal mag.:

Magnification:	Preset name:	Image Shift (x,y):	Dimension:	Binning:	Camera Dose (e/camera pixel/s):	Exposure Time (ms):	Specimen Dose (e/A^2):	Spot Size:	Defocus (m):	C2 (um):	frame saving:
1550	gr	Aligned	1024x1024(Taken with Ceta)	4	?	250	n/a	4	0.0	70	n/a
940	sq*	Aligned	927x959	4	8	400	n/a	9	-2e-4	70	no
3600	sq*	Aligned	927x959	4	8	100	n/a	9	-5e-5	70	no
8700	hl	Aligned	927x959	4	8	300	0.008	7	-6e-5	70	no
22500	fc	0,0	924x924	2	>=8*	500	8.8	7	-2.9e-7	70	no
22500	fm*	0,0	n/a	n/a	high	n/a	high	5	n/a	70	no
130000	fa	0,0	927x959	4	>=8*	200	3.5	6	-1.5e-6	70	no
130000	en	0,0	3710x3838	1	8 e/physical pixel/s and ~1.3 um on the specimen	6000	43	7	(-1e-6 to -2e-6)	70	200 ms/frame

• Note about sq preset: Depending on the alignment of the scope in EF-TEM, the beam at lowest SA mag may not expand enough without losing zero-loss peak. We have tried on different scopes either using a mag in LM mode (940x) and another at higher mag SA mode (3600x). Each has pro and con. 940x has more image-shift hysteresis but gives a full grid square. 3600x has a rather small imaging area that makes determining exposed area difficult. For 3600x, we may use a medium mag stitch application similar to MSI-PP-Stitch.

• Note: This experiment has an additional preset fm used for melting the ice exclusively. The screen goes down during the melt, therefore, it does not cause radiation damage on the camera. Do not attempt to acquire an image with this preset.

• Note: If the correlation or power spectrum don't work well enough for the standard Camera Dose Rate of 10 e/pixel/s for fc and fa, you may increase it more so not to have to use longer exposure time. The recommended value in literature is meant to give highest DQE at all sampling frequencies. Using a higher camera dose by 50% will certainly not damage the sensor.

• Note: no defocal pair was acquired in this experiment. Therefore, no ef preset.

Calibrations¶

If this is the first time Leginon is used under the scope-camera combination at the preset magnifications, complete all the calibrations listed in the chapter on calibrations. Calibrations are extremely important to operate the TEM in a consistent manner. As long as the calibrations are stable, later users do not need to perform them. Most calibrations are HT and magnification dependent. Therefore, new calibration may be required if your presets use different HT and/or magnification from all previous users.

---

< Summary of MSI applications | Initial MSI application preferences >

---