Aug 29, 2024

Public workspaceProtocol for the preparation of Wings and Wing Interference Pattern Imaging from ethanol preserved specimens

DOI
dx.doi.org/10.17504/protocols.io.bp2l6xyy1lqe/v1
  • 1Leibniz Institute for the Analysis of Biodiversity Change, Museum Koenig Bonn
  • Saskia B. Bartsch: Preparation of the first draft of the protocol
  • Jonathan Vogel: Adjusting the protocol with tweaks, preparation of final protocol
  • Anne Ankermann: Contributing the automated clean-up with Python in Section 3
  • Christoph Mayer: Supervision
  • Ralph S. Peters: Supervision
Jonathan Vogel
Icon indicating open access to content
QR code linking to this content
DOI: dx.doi.org/10.17504/protocols.io.bp2l6xyy1lqe/v1
Protocol Citation: Saskia B. Bartsch, Jonathan Vogel, Anne Ankermann, Christoph Mayer, Ralph S. Peters 2024. Protocol for the preparation of Wings and Wing Interference Pattern Imaging from ethanol preserved specimens. protocols.io https://dx.doi.org/10.17504/protocols.io.bp2l6xyy1lqe/v1
Manuscript citation:
Jonathan Vogel, Mattias Forshage, Saskia B. Bartsch, Anne Ankermann, Christoph Mayer, Pia von Falkenhausen, Vera Rduch, Björn Müller, Christoph Braun, Hans-Joachim Krammer, Ralph S. Peters () Integrative characterization of the Northwestern European species of Anacharis Dalman, 1823 (Hymenoptera: Cynipoidea: Figitidae) with the description of three new species. Journal of Hymenoptera Research
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: In development
We have been using this protocol yielding satisfactory results in many cases. However, there is most definitely room for improvement.
Created: December 07, 2023
Last Modified: August 29, 2024
Protocol Integer ID: 91948
Keywords: Wing interference patterns, WIPs, Flying insects, Preservation, Small insects
Funders Acknowledgement:
Federal Ministry of Education and Research of Germany
Grant ID: 16LI1901A
Disclaimer
DISCLAIMER – FOR INFORMATIONAL PURPOSES ONLY; USE AT YOUR OWN RISK

The protocol content here is for informational purposes only and does not constitute legal, medical, clinical, or safety advice, or otherwise; content added to protocols.io is not peer reviewed and may not have undergone a formal approval of any kind. Information presented in this protocol should not substitute for independent professional judgment, advice, diagnosis, or treatment. Any action you take or refrain from taking using or relying upon the information presented here is strictly at your own risk. You agree that neither the Company nor any of the authors, contributors, administrators, or anyone else associated with protocols.io, can be held responsible for your use of the information contained in or linked to this protocol or any of our Sites/Apps and Services.
Abstract
This protocol aims to provide standardisation for the necessary steps, specifically to a) produce standardised, artifact-free and high-quality WIP images, b) secure long-term-storage of the specimens, c) make the prepared wings re-traceable to the specimens they have been taken from d) secure reproducibility of the imaging even decades after preparation (i.e., easy cleaning of the wing specimens prior to imaging), and to e) allow reproduction of the protocol with minimal financial resources. In all these aspects, there is undeniably room for improvement.
Image Attribution
Preparation of Anacharis wing interference patterns: Before and After. By Jonathan Vogel.
Materials
Section 1 - Slide preparation
  • Stereo microscope
  • Ethanol-preserved specimens
  • Size 0 insect pin
  • Small petri dish or staining block
  • Microscopic slide (1 per specimen)
  • Coverslip (1 per specimen)
  • Ethanol (few drops per specimen)
  • Microfiber cloth
  • Fine Forceps
  • Clear nail polish
  • Small 00- or 000-type brush
  • Download Alignment sheet.pdfAlignment sheet.pdf13KB

Section 2 - Imaging
  • Zeiss Discovery.V20 SteREO microscope (60mm focal width lens)
  • AxioCam HR R3-ZEN pro 2012 software (CARL ZEISS AG)
  • Custom ring light and dome system (see Kawada and Buffington (2016))
  • Black Cardboard

Section 3 - Image Processing (req. software)
  • GNU Image Manipulation Program (GIMP development team)
  • Python package OpenCV (cv2, cf. Bradski 2000)
  • Python package imutils v0.5.4 (cf. Rosebrock 2015)
  • Python package NumPy (cf. Harris et al. 2020)
  • Python package Pandas (Pandas Development Team 2020)

Before start
The protocol is tested for both DNA-lysed and unlysed samples in high-% EtOH. While both scenarios worked, we found it a little easier to detach the wings (step 1) from the lysed samples.
Slide preparation
Slide preparation
5m
5m
Using the microscope, transfer the ethanol-preserved specimen into a small petri dish or staining block filled with ethanol.
Dissect the right fore- and hindwings of the specimen by grabbing the base of the forewing with the fine forceps and spinning the wing while pushing the specimen into the opposite direction until the wing detaches.
Repeat with the hindwing.
1m
Transfer the remaining specimen back into its labelled tube or proceed with mounting the specimen.
Clean a microscopic slide with ethanol and a microfiber cloth.
Position the printed alignment sheet (Download Alignment sheet.pdfAlignment sheet.pdf13KB ) underneath the slide.
30s
Place a small drop of ethanol in the centre of the slide.
Under the microscope, using forceps, transfer the fore- and hindwings into the ethanol drop.
30s
Align the fore- and hindwings according to the alignment sheet (see step 5) for standardised positioning and easier comparison. Make sure the fore- and hindwings do not overlap and are stretched out as good as possible. If too little ethanol is used on the slide, the wings are prone to rip apart when moved.
1m
Critical
Gently place the coverslip on top to protect the wings from dust and damage for long-time storage.
Add an additional drop of ethanol on the coverslip's edge and gently push the slip down with the forceps right above the wings to push potential air bubbles to the side. Avoid pushing too hard to not break the coverslip and not adhese the glasses. The latter is important as the WIPs will not show otherwise.
Allow the ethanol to evaporate.
30s
Critical
Gently apply a small amount of clear nail polish (enough to stabilize the slip but without it reaching the wings) to the coverslip edges using a small 00- or 000-type brush and let the nail polish harden.
30s
Label the slide to link the wings to the corresponding specimen.
1m
Imaging
Imaging
See Materials for hard- and software used and prepare accordingly.
Configure the camera settings:
  • Exposure time: 100ms
  • Saturation: 20 (maximum)
  • Contrast: Maximum
  • Magnification: 34.5X
  • Colour mode: RGB
  • Histogram: Best fit
Put the slide with the wings on the ring light, the black painted cardboard behind it and the dome on top
Bring the wing in focus, adjust white balance (Auto) and snapshot the image
Apply a scalebar to the image and export the image in a file format of choice
Image Processing
Image Processing
In GIMP, adjust luminance values by navigating to Colours/Levels. A window will pop up. Drag the min input level slider (black triangle) to the lowest visible margin of the input level curve and drag the max input level slider (white triangle) to the highest visible margin of the input level curve (see image). The visibility of the curve can be improved by dragging the window to full screen.
Adjust the colour levels (luminance values) by dragging the black and white triangle closer to the input curve.

The python scripts of the following steps can be accessed via https://github.com/aankm/WIP_auto-img.
Isolate the wings: In order to remove extraneous signal from the images, isolate the wings from the background. As is, the script will save the forewing and hindwing for each individual in separate files. The two wings are processed separately and recombined to a final image in a standardised way after the following sub-steps.
Note: To maintain the wings’ orientation to each other, combine both contours.
-> 01-extract_foreground.py
Create a mask from the contours detected in the image with the edge detection function cv2.Canny() by overlaying the edges with small circles of the same solid colour (grey). The size of the circles should be selected so that all edges within the desired foreground area connect into one shape. Fill all areas outside this shape with a solid colour. If the background is too noisy, this step may not work as desired.
Find all contours (read: continuous objects) within the mask using cv2.findContours(), and sort them by area size in decreasing order.
Fill the largest contour of the mask (the forewing) in a different solid colour (white). Use this mask to black out all areas of a copy of the original image that are not part of the foreground object.
Repeat the preceding step for the second largest contour (the hindwing).
Note: If the forewing and hindwing areas intersect, or one part is missing (due to lost or generally missing wing as in, for example, Diptera), make these steps conditional on the size of the contours.
If magnification and resolution differ within the series of images: Re-size images to put them to scale.
Optional
Standardize orientation: Rotate the images so that the back margin of the forewing and the front margin of the hindwing were horizontal. The rotation function was customised from an OpenCV interactive grabcut algorithm available at https://github.com/opencv/opencv/blob/master/samples/python/grabcut.py The program opens a new window and prompts the user to draw a line in the image indicating the edge that should be horizontally aligned. The difference between the angle of this line and the 0° horizontal is calculated and the image rotated accordingly. Note that it is important to keep the direction (start and end points) of the guiding line consistent. Use any two consistent landmarks across the dataset.
-> 02-interactive_rotation.py
Crop isolated and rotated images to their minimum possible size and combine the fore- and hindwings into one image.
-> 03-crop_combine.py
Create a mask from the contours detected in the image with the edge detection function cv2.Canny() by overlaying the edges with small circles of the same solid colour (grey). The size of the circles should be selected so that all edges within the desired foreground area connect into one shape. Fill all areas outside this shape with a solid colour.
Find all contours (read: continuous objects) within the mask using cv2.findContours(). There should be one contour in each image, since the wings were already isolated.
Find the minimum dimensions of the wing contours (cv2.boundingRect()), and crop the image to the corresponding coordinates.
Using the bounding rectangle dimensions, place the wings into a blank background with specified distance from the edges and each other. A standardised scale can be placed into the images.
If necessary, adjust the brightness of the images to the same level. Do this after removing the background and exclude black pixels from the calculation.
-> 04-adjust_brightness.py
Optional
Protocol references
- Bradski G. 2000. The OpenCV Library. Dr. Dobb’s Journal of Software Tools 25(11), p. 120-123.
- Harris CR, Millman KJ, van der Walt SJ, Gommers R, Virtanen P, Cournapeau D, Wieser E, Taylor J, Berg S, Smith NJ, Kern R. 2020. Array programming with NumPy. Nature 585(7825), p. 357-362.
- Kawada R, Buffington ML (2016) A Scalable and Modular Dome Illumination System for Scientific Microphotography on a Budget. Esposito A (Ed.). PLOS ONE 11 (5): e0153426. https://doi.org/10.1371/journal.pone.0153426
- Pandas Development Team (McKinney et al.). 2020. pandas-dev/pandas: Pandas (v1.3.3). DOI 10.5281/zenodo.3509134
- Rosebrock A. 2015. Imutils: A series of OpenCV convenience functions. https://pyimagesearch.com/2015/02/02/just-open-sourced-personal-imutils-package-series-opencv-convenience-functions/ - https://github.com/PyImageSearch/imutils
- Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, Burovski E, Peterson P, Weckesser W, Bright J, van der Walt SJ, Brett M, Wilson J, Millman, KJ, Mayorov N, Nelson ARJ, Jones E, Kern R, Larson E, Carey CJ, Polat İ, Yu F, Moore EW, VanderPlas J, Laxalde D, Perktold J, Cimrman R, Henrikson I, Quintero EA, Harris CR, Archibald AM, Ribeiro AH, Pedregosa F, van Mulbregt P, SciPy 1.0 Contributors. 2020. SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python. Nature Methods, 17(3), 261-272.