Data from: Chemical response to insect herbivory preserved in fossil leaves

This dataset compiles the images, data and scripts used in the preparation of the manuscript “Chemical response to insect herbivory preserved in fossil leaves” by Benjamin Adroit and co-authors. Metadata are documented in the filenames, including specimen numbers and data type. Detailed descriptions of the methods used to acquire and generate the data are provided in the manuscript. Necessary additional information are provided as Usage notes. The dataset includes: The images used in the figure panels, named according to their corresponding panel numbers in the manuscript. The raw photoluminescence emission and excitation spectra (GMH Pb-107_PL_emission_spectra.txt & GMH Pb-107_PL_excitation_spectrum.txt). The raw individual photoluminescence images, identified by the following filename pattern: "SpecimenNumber_PL_image_xIlluminationWavelength_fFilterRange_ExposureTime.tif". Usage notes: Unlike optical photographs, these images cannot be open and visualized as easily because they are 16-bit grayscale images and are low-intensity luminescence images collected with long exposure times; the luminescence signal can be so low that only the first hundreds of levels of grey out the 65,535 available are used, therefore making the image look all dark when opened using classical computer viewers (they read RGB or 8-bit encoding, not 16-bit). Consequently, these images are visualized and processed using viewers/software that allows for intensity adjustments of such 16-bit images. We recommend usage of the freeware ImageJ/FIJI, which is also used to generate false color RGB overlays by combining 3 of the raw images. In ImageJ/FIJI, individual .tif photoluminescence images (one by one or by selecting several at once) can be opened by simply dragging and dropping the file(s) into ImageJ/Fiji. Intensity adjustments can then be made by using the ">Image > Adjust > Brightness/contrast" tool, either automatically by using the “auto” button, or manually by setting minimum and maximum values using the slider thumbs or using the “set” button. False colour RGB overlays are also generated In ImageJ/FIJI. Following the above guidelines, this requires to open at least 3 individual .tif images. The images are then gathered into a stack using the ">Image > Stacks > Images to Stack" tool. (Note that intensity adjustment at this stage is based on the minimum and maximum values of all the opened images, so that some might appear all dark in the stack, this is normal and will be changed later). In some versions of ImageJ/FIJI, the number of images present in the stack is considered to be “Slices (z)” in the “>Image >Properties…” module, but it should be in “Channels (c)” for the generation of our RGB overlays, so these numbers need to be inverted. The RGB overlay can now be generated by using the “>Image >Color >Make Composite” tool (Display Mode “Composite”); the first image in the stack is now given a red (R) scale instead of a grey scale, the second is given a green (G) scale and the third a blue (B) scale (Note that up to 7 images (channels) can be assembled into such a “composite”; in that case the 4th image will be given a grey scale, the 5th image a cyan scale, the 6th a magenta scale and the 7th a yellow scale. Note also that generating images overlaying more than just 3 channels is possible but makes image interpretation much harder afterwards and is therefore not recommended). Once the “Composite” is created, the “> Brightness/contrast" tool can be used as described above to adjust the intensity of the different images (Note that intensity adjustments will affect the image/channel presently selected within the stack, such that adjustments need to be made for each image). The color scales can be changed using the “>Image >Color >Channels Tool…” tool, which allows for the display of all or some of the channels present in the stack and using the “More »” button to change the color scale used for the image/channel presently selected within the stack. The raw XAS data (MNT-NEL-Plante-001_XAS_gold_spectrum.nxs, MNT-NEL-Plante-001_XAS_red_spectrum.nxs & XAS_Se_foil_spectrum.nxs). The raw XRF datasets (MNT-NEL-Plante-001_XRF_data_overview.h5 & MNT-NEL-Plante-001_XRF_data_closeup.h5), and the PyMCA energy calibration and ROIs files used to identifiy and extract the individual elemental maps. Usage notes: These datasets (.h5 file) can be opened using the PyMCA data-analysis freeware (which was used in this work to generate all the data used in this work). Depending on the PyMCA version, these files may need to be opened using the “Load and show as 1D stack” option. The individual elemental maps as extracted with the PyMCA freeware using integrated intensities from the Kα peaks. Usage notes: Similarly to the photoluminescence images, raw XRF counts are decimal and the resulting images are 16-bit grayscale images that can look all dark when opened using classical computer viewers. They can also be opened by simply dragging and dropping the file(s) into ImageJ/Fiji. Intensity adjustments can then be made by using the ">Image > Adjust > Brightness/contrast" tool, either automatically by using the “auto” button, or manually by setting minimum and maximum values using the slider thumbs or using the “set” button. And images can be combined into false color RGB overlays following the procedure given above for photoluminescence images. The average decomposed XRF spectra extracted from areas of interest and decomposed using PyMCA, and the PyMCA configuration file used for energy calibration and spectral decomposition. The R_scripts used to plot and/or process the spectrocopy data, and to generate some of the associated figure panels.