Disclaimer

pyTEM was developed for use on the BASF SE TEM Laboratory's Talos F200i transmission electron microscope. The authors provide no guarantee that the software will function as intended, either in part or in whole, on any other microscope installation.

pyTEM is available under the MIT license here.

About

pyTEM is a high-level scripting interface enabling the user-friendly control of, and automated data acquisition on, Thermo Fisher Scientific and FEI microscopes from a pure Python environment. Bolted directly on top of a COM interface, pyTEM is a Python wrapper for, and extension of, the prerequisite Thermo Fisher Scientific / FEI scripting and advanced scripting interfaces.

While it may depend on your microscope installation, pyTEM will likely need to be run on a microscope control computer with the prerequisite Thermo Fisher Scientific / FEI scripting and advanced scripting interfaces installed and properly configured. For detailed information regarding your microscope's scripting capabilities, please refer to the documentation accompanying your microscope or contact to your microscope supplier.

In addition to the main scripting interface, pyTEM ships with various scripts. Besides being useful in-and-of-themselves, these scripts demonstrate how to interface with and control the microscopes using pyTEM. A list of available scripts can be found below. These scripts are included as part of pyTEM for the sake of consolidating the TEM Laboratory's scripting efforts.

Supported Python versions: 3.8.

Contribution & Contact Info

pyTEM is developed and maintained by the TEM Microscopy Laboratory at BASF SE in Ludwigshafen, Germany. If you have any questions about the pyTEM project or would like to contribute or collaborate, please contact Philipp Müller at [email protected].

Issues should be reported to the issues board here.

Interface

pyTEM is a microscope scripting interface. This means that you can issue commands to, and receive microscope data from, compatible microscopes by calling pyTEM functions. This is not a complete interface in that it does not provide access to all the microscope's functionality. However, it provides access to all fundamental microscope functions as well as some more advanced functions which were required for the development of one or more pyTEM scripts.

pyTEM aims to be more user-friendly than the underlying Fisher Scientific / FEI scripting and advanced scripting interfaces. To this end:

• pyTEM functions return only built-in data types or instances of useful, simple classes (no pointers).
• pyTEM accepts input and returns results in user-friendly units. For example, stage position is set in microns/degrees, and tilt speed in degrees-per-second.
• pyTEM provides many additional functions not directly available through the underlying interface. For example, tilt-while-acquiring, metadata evaluation, and save-to-file functionality.
• pyTEM is open source and licensed such that users can make any changes or modifications required to suit their own installation.

Finally, pyTEM is a good starting place for those interested in learning how to control their microscope from a pure Python environment.

pyTEM controls are divided across Python mixins. This simplified class-diagram was built to help articulate pyTEM's architecture. Notice that some mixins include other mixins, and a pyTEM Interface includes all mixins along with some interface-level controls.

Controls

Acquisition Controls

Microscope acquisition controls including acquisition() and acquisition_series() functions. By means of multitasking, both functions offer beam-blanker optimization and acquire-while-tilting functionality.

These functions return pyTEM Acquistion and AcquisitionSeries objects, respectively. Both the Acquistion and AcquisitionSeries classes provide helpful methods for further image and metadata manipulation as well as save-to-file functionality.

Acquisition Controls Example

from pyTEM.Interface import Interface
from pathlib import Path


my_microscope = Interface()

# Get a list of available cameras.
available_cameras = my_microscope.get_available_cameras()

if len(available_cameras) > 0:
    # Let's see what each camera can do...
    for camera in available_cameras:
        my_microscope.print_camera_capabilities(camera_name=camera)

    # Perform a single blanker-optimized acquisition using the first available camera.
    acq = my_microscope.acquisition(camera_name=available_cameras[0], exposure_time=1, 
                                    sampling='4k', blanker_optimization=True)

    # Downsample the acquisition (bilinear decimation by a factor of 2).
    acq.downsample()
    
    # Display a pop-up with the results of our acquisition.
    acq.show_image()

    # Save the acquisition to file.
    downloads_path = str(Path.home() / "Downloads")
    acq.save_to_file(out_file=downloads_path + "/test_acq.tif")

else:
    print("No available cameras!")

Magnification Controls

When in imaging mode, get and set both TEM and STEM magnification. When in diffraction mode, get and set camera-length.

Magnification Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

# Make sure we are in TEM imaging mode.
my_microscope.set_mode(new_mode="TEM")
my_microscope.set_projection_mode(new_projection_mode="imaging")

# Print out the current magnification.
current_magnification = my_microscope.get_magnification()
print("Current magnification: " + str(current_magnification) + "x Zoom")

# Print a list of available magnifications.
my_microscope.print_available_magnifications()

# TEM magnification is set by index, let's increase the magnification by three notches.
current_magnification_index = my_microscope.get_magnification_index()
my_microscope.set_tem_magnification(new_magnification_index=current_magnification_index + 3)

# And decrease it back down by one notch.
my_microscope.shift_tem_magnification(magnification_shift=-1)

Image and Beam Shift Controls

Get and set both image and beam shift.

Image and Beam Shift Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

# Print out the current image shift.
u = my_microscope.get_image_shift()
print("Current image shift in the x-direction: " + str(u[0]))
print("Current image shift in the y-direction: " + str(u[1]))

# Print out the current beam shift.
v = my_microscope.get_beam_shift()
print("\nCurrent beam shift in the x-direction: " + str(v[0]))
print("Current beam shift in the y-direction: " + str(v[1]))

# Shift the image 2 microns to the right, and 3 microns up.
my_microscope.set_image_shift(x=u[0] + 2, y=u[1] + 3)

# Move the beam shift to (-10 um, 5 um).
my_microscope.set_beam_shift(x=-10, y=5)

# Print out the new image shift.
u = my_microscope.get_image_shift()
print("\nNew image shift in the x-direction: " + str(u[0]))
print("New image shift in the y-direction: " + str(u[1]))

# Print out the new beam shift.
v = my_microscope.get_beam_shift()
print("\nNew beam shift in the x-direction: " + str(v[0]))
print("New beam shift in the y-direction: " + str(v[1]))

# Zero both image and beam shift.
my_microscope.zero_shifts()

Mode Controls

Microscope mode controls, including getters and setters for instrument, illumination, and projection mode.

Mode Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

# Make sure we are in TEM imaging mode.
my_microscope.set_mode(new_mode="TEM")
my_microscope.set_projection_mode(new_projection_mode="imaging")

# Print out the current projection submode.
my_microscope.print_projection_submode()

# Print out the current illumination mode.
print("\nCurrent illumination mode: " + my_microscope.get_illumination_mode())
if my_microscope.get_illumination_mode() == "microprobe":
    print("This mode provides nearly parallel illumination at the cost of a larger probe size.")
else:
    print("Use this mode to get a small convergent electron beam.")
    
# Switch to STEM mode.
my_microscope.set_mode(new_mode="STEM")

Screen Controls

Insert and retract the FluCam screen.

Screen Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

# Make sure the FluCam screen is inserted.
if my_microscope.get_screen_position() == "retracted":
    my_microscope.insert_screen()

Beam Blanker Controls

Blank and un-blank the electron beam.

Beam Blanker Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

if my_microscope.beam_is_blank():
    print("The beam is blanked... un-blanking beam...")
    my_microscope.unblank_beam()
else:
    print("The beam is un-blanked... blanking beam...")
    my_microscope.blank_beam()

Stage Controls

Microscope stage controls, including those to get and set the stage position, monitor the stage status, and inspect the current specimen-holder's capabilities.

Stage Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

# Print out the stage's current status.
my_microscope.print_stage_status()

# Reset the microscope stage to the home position.
if not my_microscope.stage_is_home():
    my_microscope.reset_stage_position()
    
# Update the stage position along the x, y, and alpha-tilt axes. Move at half speed.
my_microscope.set_stage_position(x=5, y=10, alpha=-45, speed=0.5)
 
# Print out the current stage position.
my_microscope.print_stage_position()

Vacuum Controls

Vacuum system controls, including those to read gauge pressures, monitor the vacuum in the column, and control the column valve.

Vacuum Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

# Print out the current status of the vacuum system.
my_microscope.print_vacuum_status()

# Print out vacuum info in a table-like format.
my_microscope.print_vacuum_info()

# Check if the column is under vacuum.
if my_microscope.column_under_vacuum():
    print("The column is currently under vacuum; it is safe to open the column valve.")
else:
    print("Column vacuum is insufficient to open the column valve.")
    
# Make sure the column valve is closed.
if my_microscope.get_column_valve_position() == "open":
    my_microscope.close_column_valve()

Interface-Level Controls

Additionally, some pyTEM functions are defined within the Interface class itself (rather than included from one of the other mixins). Often, these functions utilize functions from the included mixins.

Interface-Level Controls Example

from pyTEM.Interface import Interface
my_microscope = Interface()

# Normalize all lenses.
my_microscope.normalize()

# Prepare for holder removal.
my_microscope.prepare_for_holder_removal()

# Return the microscope to a safe state.
my_microscope.make_safe()

Limited Interfaces

As we have seen, a pyTEM interface can be established by creating an instance of the pyTEM.Interface class. However, one can also create interfaces that support a limited subset of pyTEM functions. For example, we can create a pyTEM interface with only stage controls:

from pyTEM.lib.mixins.StageMixin import StageInterface
stage_interface = StageInterface()

Using stage_interface, we can perform the same series of stage commands performed above:

# Print out the stage's current status.
stage_interface.print_stage_status()

# Reset the microscope stage to the home position.
if stage_interface.stage_is_home():
    pass
else:
    stage_interface.reset_stage_position()

# Update the x, y, and alpha stage positions. Move at half speed.
stage_interface.set_stage_position(x=5, y=10, alpha=-45, speed=0.5)

# Print out the current stage position.
stage_interface.print_stage_position()

This technique can be useful when speed is an issue, or when it's best-practice to restrict unnecessary control. Custom interfaces can be created by including only the required mixins. For example, if you need stage and beam-blanker controls but nothing else, you may want to use an instance of the following limited interface:

import comtypes.client as cc

from pyTEM.lib.mixins.StageMixin import StageMixin
from pyTEM.lib.mixins.BeamBlankerMixin import BeamBlankerMixin


class StageBeamBlankerInterface(StageMixin, BeamBlankerMixin):
    """
    A microscope interface with only stage and beam blanker controls.
    """

    def __init__(self):
        try:
            self._tem = cc.CreateObject("TEMScripting.Instrument")
        except OSError as e:
            print("Unable to connect to the microscope.")
            raise e

Automated Image Alignment with Hyperspy

Several pyTEM functions and scripts use Hyperspy's estimate_shift2D() function to estimate the pixel offset between images. This function uses a phase correlation algorithm based on the following paper:

Schaffer B, Grogger W, Kothleitner G. Automated spatial drift correction for EFTEM image series. Ultramicroscopy. 
    2004 Dec;102(1):27-36. doi: 10.1016/j.ultramic.2004.08.003. PMID: 15556698.

Scripts

pyTEM scripts are sequences of commands that perform useful data acquisitions, initialize or return the microscope to some pre-defined state, or achieve some other common task. They, often heavily, rely on the pyTEM Interface or other pyTEM library functions. pyTEM scripts are distributed with, and automatically installed alongside, pyTEM itself.

pyTEM scripts are run from the command line. For example:

micro_ed --verbose

To view script usage, use the --help option:

micro_ed --help

Often, pyTEM scripts utilize custom Tkinter UIs to simplify and streamline IO.

Since all pyTEM scripts utilize pyTEM itself, running pyTEM scripts requires all of pyTEM along with its prerequisites and dependencies.

micro_ed

micro_ed is BASF's micro-crystal electron diffraction (MicroED) automated imaging script. MicroED allows for the fast, high-resolution 3D structure determination of small chemical compounds and biological macromolecules. More on MicroED here.

micro_ed achieves automated image alignment by computing the image deviation during a preparatory tilt sequence and then applying a compensatory image shift during the main acquisition sequence. This automated image alignment functionality is optional. micro_ed only collects the data, it does not analyse it.

micro_ed results can be saved as a single multi-image stack file or as multiple single-image files. Since performing a MicroED acquisition sequence requires some actions that aren't easily automated (such as selecting suitable particles and setting the eucentric height), a series of Tkinter message boxes guide the user through those steps requiring manual interaction.

align_images

align_images allows the user to load some images from file (possibly from a single multi-image stack file or possibly from multiple single-image files), align the images, and then save the results back to file as a single multi-image stack.

Colour images are converted to 8bit unsigned greyscale prior to alignment.

bulk_carbon_analysis

Given 16 natural light-micrographs of a bulk carbon sample sandwiched between polarizers of varying cross, produce both anisotropy and orientation maps.

Colour images are converted to 8bit unsigned greyscale prior to analysis.

This script uses the technique explained in the following paper:

Gillard, Adrien & Couégnat, Guillaume & Caty, O. & Allemand, Alexandre & P, Weisbecker & Vignoles, Gerard. (2015). 
    A quantitative, space-resolved method for optical anisotropy estimation in bulk carbons. Carbon. 91. 423-435. 
    10.1016/j.carbon.2015.05.005.

controller

Allows users to control their microscope with an Xbox controller. This is especially helpful for microscope operators working off-site.

Installation

Because pyTEM is often required on microscope control machines which lack internet connectivity, pyTEM is not listed on the Python Package Index, nor anywhere else. Rather, we provide a wheel file in /dist and the following offline install instructions.

Offline Install Instructions

Part 1. On a system with internet access:

1. Download requirements.txt.
2. Download the required dependencies with pip pip download -d ./pytem_dependencies -r requirements.txt.
3. Download the pyTEM wheel from /dist.

Transfer requirements.txt, the entire pytem_dependencies folder you just created, and the pyTEM wheel to the offline machine (any directory is fine; most opt for current user's Downloads directory).

Part 2. On the offline system:

1. Ensure both pip and wheel are already installed.
2. Install the required dependencies with pip: pip install --no-index --find-links ./pytem_dependencies -r requirements.txt
3. Install pyTEM itself with pip. Example: pip install pyTEM-0.1.0-py3-none-any.whl

If you run into any problems installing the required dependencies, check that you are using the same version of pip and wheel on both the online and offline systems.

In case you need to build your own custom pyTEM wheel

1. Install wheel with pip install wheel.
2. Download the whole pyTEM project directory.
3. Navigate to the pyTEM folder.
4. Run setup.py with the bdist_wheel setuptools command. Example: python setup.py bdist_wheel

Authorship

pyTEM is developed and maintained by the TEM microscopy laboratory at BASF SE in Ludwigshafen, Germany. The initial development was performed by RISE (Research Internships in Science and Engineering) Interns from North America. More on the RISE program here.

Meagan Jennings (Sept - Dec 2021)

Hometown:

Baltimore, Maryland, USA

Contributions:

• Figured out how to interface with and control the microscope from a pure Python environment.
• developed TEMPackage, the predecessor to pyTEM. View the project as it existed at the time of Meagan's final commit here.
• Developed microED_Tilt_Series, the predecessor to pyTEM's micro_ed script. View the project on GitLab here.
• Wrote the original TEM Scripting Guide, which can be found in /docs.

Michael Luciuk (May - Aug 2022)

Hometown:

Saskatoon, Saskatchewan, Canada

Contributions:

• Refactored the original TEMPackage module into the pyTEM library that we know and love today.
• Refactored the original microED_Tilt_Series script into the micro_ed script that we know and love today.
• Updated and improved both pyTEM and micro_ed.
• Developed the align_images script and did some initial work on the bulk_carbon_analysis script.

You can view the pyTEM project as it existed at the time of Michael's final commit here.