Join GitHub today
Spot intensity detection
Fetching latest commit…
Cannot retrieve the latest commit at this time.
SID (Spot Intensity Detector) software =================================== SID is a MATLAB program for detecting spot intensities in microscopy images in 2D or 3D, in up to four channels. This software is licensed for use under the GNU GPL v3 license. See LICENSE file for details. Installation ------------ No installation is necessary, simply copy the source code to the directory of your choice. Enter that directory within MATLAB (use directory toolbar at top) and then enter commands. Usage ----- A spot detection experiment can be processed using SID’s home console. To access the home console: sidRun This will display a GUI, from which all tasks can be processed – no more commands are required. The home console allows creation of a set of new runs (or a jobset). 1. Creating new runs Pressing ‘New Run’ will display a GUI to allow modification of spot detection parameters. Click 'Select directory' to choose a directory to search for image files. Select the images to analyse by highlighting them in the list, then either click ‘Add’ to include the full image, or ‘Add + crop’ to allow cropping of one or more ROIs. When selecting ROIs, click ‘Add ROI’, then double-click after drawing each to save. Click ‘Finish’ when all ROIs have been selected for that image. After ROI selection, configure channel and spot detection options. Select a channel in which to detect spots, then select in which additional channels intensities should be measured. The default spot detection options were tailored to detection of kinetochores, so minimum and maximum spots may need to be changed on a case-by-case basis. The value for intensity restriction is a value between 0 and 1, and should be increased to include dimmer spots. Choose a name for the jobset. Finally, click ’Save' to allow further processing in the home console. Clicking ‘Load Run’ will alternatively allow selection of a previously-created jobset. Once a jobset has been created or loaded, its file path is shown below the ‘New Run’ and ‘Load Run’ buttons. The ‘Edit’ and ‘Run’ buttons then also become available. To edit this jobset’s detection parameters, click ‘Edit’. To run spot detection for this jobset, click ‘Run’. After spot detection is complete, you will find a file named something like ‘siddetection001_exptname_imagefile.mat', where exptname is replaced with the name of the jobset mat-file, and imagefile is replaced with the name of the image to which this result is associated. To run multiple jobsets at once, click ‘Multi-run’. This will display a GUI to allow creation and loading of jobsets. Upon creation/loading, jobsets are listed below. If a jobset is already loaded into the home console, it will be listed here. To remove a jobset from the list, highlight the given jobset and click ‘Remove’. Once all jobsets for processing are listed, click ‘Run’. Click ‘Close’ once completed. 2. Manual filtering of data Once the jobset loaded into the home console has been run, manual filtering of both spots and cells can be done. Clicking ‘Spot filtering’ will display a GUI showing a grid of all the spots detected for a given image. Click on erroneous spots to deselect them, so that only spots with green (red) borders are included (ignored) in analysis. Click ’Invert’ to invert all green borders to red, and vice versa. Click ‘Next’ and ’Prev’ to move between images, and click ‘Finish’ when manual filtering is complete. N.B. No data is deleted. Ignored spots can be re-selected at a later time. Clicking ‘Cell filtering’ will display a GUI showing a z-projection of each image from the jobset. Dot plots will also be shown to demonstrate the measured intensities in each channel for this image (red circles), relative to the data for the full jobset (grey). Selecting either ‘Keep’ or ‘Discard’ will determine whether this cell is included in analysis. This tool can be used to remove either cells visibly showing an unwanted phenotype, or cells whose intensities are outliers of the full jobset population. Click ‘Finish’ when cell filtering is complete. N.B. No data is deleted. Ignored cells can be re-selected at a later time. 3. Full analysis Once a number of jobsets have been run, analysis of the data can be done by clicking ‘Full analysis’. This displays a GUI to allow selection of jobsets for multiple experimental conditions for comparison. Conditions are separated into tabs, where they can be labelled. One condition can be selected for normalisation (labelled *), so that all other conditions are plotted as a fraction of that condition. Select multiple jobsets per condition by clicking ‘Find jobset(s)’, and highlight the required jobsets. Select a control channel (one that is thought to be consistent per condition), and additional channels to be analysed, providing labels for each. Select a preferred plotting style. Click ‘Save directory:’ to select where to save analysis, and provide a filename. Click ‘Run’ when complete. Figures can be forced to remain open in MATLAB by ticking the checkbox. Multiple files are saved in the save directory, including a .mat file, and .csv files per condition, of intensity data. For each figure, an .eps file is saved. Figures show background-corrected, cell-normalised (each spot normalised to its cell’s average control channel intensity), and spot-normalised intensities (each spot normalised to its own control channel intensity). Click ‘Close’ when finished’. Bugs ---- Send bug or crash reports, and any feature requests, to email@example.com. Include: - Brief description of steps to reproduce error, - Copy of the error text, - Jobset mat-file (if available). Images themselves will be useful but should be sent via e.g. Dropbox. Credits ------- SID evolved from KiT (Armond et al. 2016, Bioinformatics. 2016;32(12):1917-9). Portions of the core spot detection code is adapted from KiT and renamed appropriately. C. Currie wrote the SID for Dummies document. The SID package contains code from various other sources. See file headers to identity sources and licenses. MATLAB requirements -------------------- MATLAB R2017a MATLAB Image Processing Toolbox 9.1 MATLAB Statistics Toolbox 9.1