documents:121005advancedimg
Differences
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
documents:121005advancedimg [2012/10/06 15:08] – kota | documents:121005advancedimg [2016/05/24 12:46] (current) – external edit 127.0.0.1 | ||
---|---|---|---|
Line 1: | Line 1: | ||
+ | ====== EMBL Advanced Imaging Course: Tracking principles and Practicals ====== | ||
+ | Lecture notes and the manual for the practical is available in the following link: | ||
+ | * [[https:// | ||
+ | |||
+ | Please download. | ||
+ | ===== Downloads and Installation for the Practical ===== | ||
+ | |||
+ | We use Fiji in the practical work. Additionally, | ||
+ | |||
+ | * [[http:// | ||
+ | * [[http:// | ||
+ | * [[http:// | ||
+ | |||
+ | If we might be able to also try | ||
+ | * [[https:// | ||
+ | |||
+ | ====== Manual: Tracking Principles / Practicals ====== | ||
+ | |||
+ | The following content is same as the PDF downloadable form the link above. | ||
+ | |||
+ | < | ||
+ | In the lecture, we will overview methods used for tracking in cell and | ||
+ | developmental biology. Focus is made on the details of generally used | ||
+ | two-step strategy namely particle detection and linking. In practical | ||
+ | works, we use several image processing and analysis tools and explore | ||
+ | tracking techniques for measuring EB3 movement in cultured cells. We try | ||
+ | manual tracking, kymograph and automated tracking. | ||
+ | |||
+ | Lecture Notes | ||
+ | ============= | ||
+ | |||
+ | Time series of digital images, usually called ‘a stack’, contains | ||
+ | temporal dynamics of position and intensity. By analyzing these | ||
+ | dynamics, we can extract numerical parameter which then enables us to | ||
+ | characterize the biological system. There are three types of dynamics. | ||
+ | |||
+ | 1. Position does not change but intensity changes over time. | ||
+ | |||
+ | 2. Position changes but the intensity does not change. | ||
+ | |||
+ | 3. Both Position and Intensity change over time. | ||
+ | |||
+ | The lecture is about the measurement of type 2, generally called object | ||
+ | tracking. Realistic situation demands type 3 *e.g.* you need to get a | ||
+ | FRAP curve out of moving organelle, but we approximate the sample as | ||
+ | type 2 as we need to learn the principle of tracking first, and type 3 | ||
+ | will be your research that asks your ability to apply the full knowledge | ||
+ | you acquired in the EMBL Advanced Imaging Course 2012, and probably some | ||
+ | technique to use scripting languages. | ||
+ | |||
+ | For recent review on object tracking in cell and developmental biology, see Meijering et. al. (2012). This article also contains an extensive list of available tracking packages. A bit old review on single particle tracking by Saxton is still an excellent textbook affording rich biophysical insights for analyzing tracking results (Saxton and Jacobson 1997). For the classic “tracking” mentioned in the talk, or the presumptive fate map, see Vogt ()1925) only if you love biology (I am sure you do). This historical article could be downloaded via Google books. Magnificent movies produced by Keller et. al. using | ||
+ | single plane illumination microscopy and extensive image processing that almost killed EMBL cluster by steaming it could be accessed through the web page of the paper in Science (Keller et al. 2008). Comparison of segmentation strategy in object tracking was published in 2001 by Cheezum et. al (2001) and has been a frequently cited paper in the field of biological object tracking. For overall review on image analysis methods see (Hamilton 2009). | ||
+ | |||
+ | Segmentation techniques | ||
+ | ----------------------- | ||
+ | |||
+ | - | ||
+ | ImageJ “manual tracker plugin”, we will do this in the practical. | ||
+ | |||
+ | - | ||
+ | |||
+ | - | ||
+ | |||
+ | - Local intensity maxima | ||
+ | |||
+ | - | ||
+ | |||
+ | - Many applications for getting sub-pixel resolution coordinates. | ||
+ | |||
+ | - You also saw this already in super-resolution microscopy talks | ||
+ | by Ricardo and Melike. | ||
+ | |||
+ | - | ||
+ | |||
+ | - | ||
+ | |||
+ | - < | ||
+ | |||
+ | - | ||
+ | < | ||
+ | |||
+ | - | ||
+ | |||
+ | - There is no tracking package that does tracking using machine | ||
+ | learning based segmentation, | ||
+ | coming days. | ||
+ | |||
+ | - Fiji plugin “Advanced Weka Segmentation” | ||
+ | < | ||
+ | |||
+ | - | ||
+ | |||
+ | - 3D tracking of macrophage-like cells. | ||
+ | See (Grabher et al. 2007) | ||
+ | |||
+ | Linking techniques | ||
+ | ------------------ | ||
+ | |||
+ | - | ||
+ | |||
+ | This is the simplest principle linking an object in one time point | ||
+ | to the other. | ||
+ | |||
+ | - | ||
+ | |||
+ | - | ||
+ | |||
+ | - | ||
+ | |||
+ | - | ||
+ | |||
+ | - (I.F. Sbalzarini and Koumoutsakos 2005) | ||
+ | |||
+ | - (Ivo F. Sbalzarini 2006) | ||
+ | |||
+ | - | ||
+ | |||
+ | - | ||
+ | |||
+ | - | ||
+ | |||
+ | Tracking Packages we do not try, but worth mentioning here. | ||
+ | ----------------------------------------------------------- | ||
+ | |||
+ | For single molecule trackers, there is a pretty new plugin called PTAj, | ||
+ | written by **Yoshiyuki Arai**. We do not try this plugin in the practical | ||
+ | but for those involved in single molecule imaging, it might be worth | ||
+ | trying it at your home institute. The plugin offers sophisticated | ||
+ | interface to review the tracking results. We all know that even though | ||
+ | the tracking is automatic, we start checking each track manually just to | ||
+ | be sure, and just to be sure… | ||
+ | |||
+ | - < | ||
+ | |||
+ | There is also a new “all-in-one” tracking plugin for ImageJ. It’s name | ||
+ | is TrackMate, written by **Jean-Yves Tinevez**. The linking algorithm | ||
+ | includes the most-up-to-date “Tracklets” and allows to track | ||
+ | merging/ | ||
+ | thoroughly, but the interface is similar to Imaris and should be easy | ||
+ | for people who are used to wizard-style parameter setting for tracking. | ||
+ | For developers, the attractive feature of this plugin is that it offers | ||
+ | interface to implement your own algorithm for segmentation and linking. | ||
+ | |||
+ | - < | ||
+ | |||
+ | Sample Sequences | ||
+ | ================ | ||
+ | |||
+ | ##Get Image Files | ||
+ | |||
+ | If capturing of EB3 image sequences were successful (I hope so!!), | ||
+ | we use them. Otherwise, we use an example image sequence of EB1 | ||
+ | labeled cells. EB1 behaves similar to EB1 and moves along | ||
+ | microtubules. A file named **eb18\_b.tif** is a sequence taken from | ||
+ | single cultured cell labeled with eb1. This sequence is accessible | ||
+ | by downloading an ImageJ plugin " | ||
+ | for installing this plugin is written in below. | ||
+ | |||
+ | Download and save the image files, the sequence you have taken or | ||
+ | the example sequence eb18\_b.tif in your local computer. We try | ||
+ | several different ways to measure movement of EB3 or EB1 protein by | ||
+ | tracking. Before tracking, examine the sequence first using stack | ||
+ | related functions. | ||
+ | |||
+ | ####Open the file in Fiji | ||
+ | |||
+ | `[File > Open...]` | ||
+ | |||
+ | ####Check dimensions of the image | ||
+ | |||
+ | `[Image > Properties...]` | ||
+ | |||
+ | …Image stacks are by default taken as a z-series and not t-series. Set Slices to 1, and Frames to appropriate size (number of frames). | ||
+ | |||
+ | ####Examine the sequence using stack tools. | ||
+ | |||
+ | Explore stack functions. | ||
+ | Start animation, Stop animation, change frame rates, | ||
+ | Manipulations…These functions are located under | ||
+ | |||
+ | `[Image > Stack >]` | ||
+ | |||
+ | ####A useful plugin for stack: Running Z-Projector | ||
+ | |||
+ | | ||
+ | | ||
+ | | ||
+ | | ||
+ | |||
+ | - < | ||
+ | |||
+ | 1. Download the plugin from above URL. | ||
+ | 2. Install the plugin using `[Plugin > Install...]` | ||
+ | 3. Restart Fiji, just to be sure with the installation. | ||
+ | 4. Try running it on EB1 or EB3 sequence. | ||
+ | |||
+ | ####A plugin for sample images | ||
+ | |||
+ | Download the plugin " | ||
+ | |||
+ | - < | ||
+ | |||
+ | Manual Tracking | ||
+ | =============== | ||
+ | |||
+ | In the simplest case, tracking can be done manually. The user can read | ||
+ | out the coordinate position of the target object directly from the | ||
+ | imaging software. In ImageJ, user can read out position coordinate | ||
+ | indicated in the status bar by placing the cross-hair pointer over the | ||
+ | object. Then coordinates can be listed in standard spreadsheet software | ||
+ | such as Microsoft Excel or R for further analysis. | ||
+ | |||
+ | An ImageJ plug-in is also freely available to assist such simple way of | ||
+ | tracking. An obvious disadvantage of the manual tracking is that the | ||
+ | mouse-clicking by the user could be erroneous, as we are still human who | ||
+ | gets tired after thousands of clicking. For such errors, measurement | ||
+ | errors can be estimated by tracking the same object several times and | ||
+ | this error could then be indicated together with the results. Otherwise, | ||
+ | automated tracking is more objective, but if you have only limited | ||
+ | number of tracks to analyze, manual tracking is a best choice before | ||
+ | start to explore complex parameter space of automated tracking setting. | ||
+ | |||
+ | The manual tracking can be assisted by an ImageJ plug-in " | ||
+ | Tracking" | ||
+ | the position where the user clicked using mouse in each frame within a | ||
+ | stack. The download site has a detailed instruction on how to use. | ||
+ | |||
+ | #### | ||
+ | |||
+ | **Manual Tracking (Manual\_Tracking.class)** is a plugin bundled | ||
+ | with Fiji. If you are using ImageJ, download the plugin file from | ||
+ | the link below. The plugin was written by **Fabrice Cordelieres**. | ||
+ | |||
+ | - < | ||
+ | |||
+ | #### | ||
+ | |||
+ | 1. Your task now is to track the movement of individual EB3 signal using the ManualTracker PlugIn. To ease detection of the signal, enhance contrast and convert the image stack from 16-bit to 8-bit. | ||
+ | |||
+ | `[Image > Adjust > Brightness and Contrast]` | ||
+ | |||
+ | and then convert to 8-bit | ||
+ | |||
+ | `[Image > Type > 8bit]` | ||
+ | |||
+ | 2. Then | ||
+ | |||
+ | | ||
+ | |||
+ | | ||
+ | |||
+ | | ||
+ | |||
+ | | ||
+ | |||
+ | You could track different particles by repeating steps between 2 | ||
+ | and 3. Results window will list the measured positions for | ||
+ | | ||
+ | |||
+ | 3. Tracked data can be saved by activating " | ||
+ | |||
+ | `[File > Save as...]` | ||
+ | |||
+ | 4. **OPTIONAL: | ||
+ | in Excel or R and calculate a histogram of velocity | ||
+ | | ||
+ | |||
+ | Kymograph | ||
+ | ========= | ||
+ | |||
+ | Kymographs are a two-dimensional time traces, where time *t* is in Y | ||
+ | axis and space along a one-dimensional contour is in *X*, and the | ||
+ | dynamical variable *F(x,t) *is visualized as an image. Kymographs | ||
+ | provide a fast and convenient way to visualize motion and dynamics in | ||
+ | microscopy images.We try measuring the speed of EB1 or EB3 movement | ||
+ | using Kymographs. | ||
+ | |||
+ | 1. Open the image stack. | ||
+ | |||
+ | 2. Contrast enhance, | ||
+ | |||
+ | `[Image > Adjust > Brightness and Contrast]` | ||
+ | |||
+ | and then convert to 8-bit | ||
+ | |||
+ | `[Image > Type > 8bit]` | ||
+ | |||
+ | 3. Do maximum intensity Z-projection | ||
+ | |||
+ | `[Image > Stacks > Z-projection]`. | ||
+ | |||
+ | 4. Choose segmented line ROI tool (to do this, right click line-ROI | ||
+ | icon in the menu bar. You will see a drop down selection). With | ||
+ | projection image generated above, trace one of the tracks in the | ||
+ | projection image. | ||
+ | |||
+ | 5. Go back to the stack (click the window and activate it - or bring it | ||
+ | to the front - ), | ||
+ | |||
+ | `[Edit > Selection > Restore Selection]` | ||
+ | |||
+ | Then | ||
+ | |||
+ | `[Image > Stacks > Reslice...]` | ||
+ | |||
+ | Don’t change parameters in the dialog window, simply OK. Resulting | ||
+ | image is the kymograph. | ||
+ | |||
+ | 6. To estimate the velocity of movement from kymograph, you simply need | ||
+ | to measure the slope of the diagonal line shown in the kymograph. | ||
+ | Since x-axis is distance and y-axis is time, if the starting point | ||
+ | of the diagonal line is $(x_1, y_1)$ and the end point of the line | ||
+ | is $(x_2, y_2)$ then the $slope = velocity$ is | ||
+ | $$velocity = (x_2 - x_1) / (y_2 - y_1)$$ At the bottom of this section, you could find a simple macro to | ||
+ | calculate this automatically from straight line selection. | ||
+ | |||
+ | ImageJ macro for measuring kymograph | ||
+ | |||
+ | To use this macro, do | ||
+ | | ||
+ | `[File > New > Script]` | ||
+ | | ||
+ | to launch script editor, then select | ||
+ | | ||
+ | `[Language > ImageJ Macro]` | ||
+ | | ||
+ | in the script editor menu, copy and paste the code above. You could also copy and paste from the following page. | ||
+ | |||
+ | - < | ||
+ | |||
+ | 7. **OPTIONAL: | ||
+ | histogram of velocity distribution. | ||
+ | |||
+ | 8. **OPTIONAL: | ||
+ | there is a good plugin that assists you in doing so efficiently. The | ||
+ | plugin was written by **Jens Rietdorf** and **Arne Seitz** | ||
+ | |||
+ | - | ||
+ | < | ||
+ | |||
+ | 9. **OPTIONAL: | ||
+ | a line, texture like streakes) use: | ||
+ | |||
+ | - | ||
+ | < | ||
+ | |||
+ | Written by myself, with valuable suggestions from Peter Lenert. | ||
+ | |||
+ | </ | ||
+ | <sxh javascript> | ||
+ | // scale and dt must be manually set | ||
+ | var xyscale=0.207; | ||
+ | var dt=0.69; //sec | ||
+ | |||
+ | // sampling ROI coordinates, | ||
+ | print(" | ||
+ | getLine(x1, y1, x2, y2, lineWidth); | ||
+ | print(" | ||
+ | +x1+" , " | ||
+ | " , " + y2 + ") "); | ||
+ | |||
+ | // calculation of speed. | ||
+ | dx=abs(x2-x1); | ||
+ | dy=abs(y2-y1); | ||
+ | dx *= xyscale; | ||
+ | dy *= dt; | ||
+ | velocity= dx/dy; | ||
+ | //output in log window. | ||
+ | print(" | ||
+ | print(" | ||
+ | </ | ||
+ | < | ||
+ | |||
+ | Automated Tracking: ParticleTracker Plugin | ||
+ | ========================================== | ||
+ | |||
+ | This plugin uses linking algorithm that involves global optimization | ||
+ | based on cost function that involves signal intensity and its second | ||
+ | order derivative. The plugin was first written as a Matlab code by **Ivo | ||
+ | Sbalzarini** and the migrated to ImageJ plugin by his group members, **Guy | ||
+ | Levy** and **Janick Cardinale**. | ||
+ | |||
+ | #### | ||
+ | |||
+ | |||
+ | The plugin could be down loaded from | ||
+ | |||
+ | - < | ||
+ | |||
+ | To install the downloaded plugin, `[Plugins > Install...]`. You | ||
+ | could also directly copy the file to plugins folder under Fiji | ||
+ | directory and do `[Help > Refresh Menus]`. | ||
+ | |||
+ | Just in case if the above plugin somehow does not work, try download | ||
+ | and use the version linked in the flowing site. | ||
+ | |||
+ | - < | ||
+ | |||
+ | #### | ||
+ | |||
+ | Start the particleTracker by | ||
+ | |||
+ | `[Plugins > Mosaic > ParticleTracker 2D/3D]`. | ||
+ | |||
+ | ####Study Dot Detection Parameter | ||
+ | |||
+ | This tracking tool has two parts. First, all dots in each frame are | ||
+ | detected, and then dots in successive frames are linked. The first | ||
+ | task then is to determine three parameters for dot detection. There | ||
+ | are three parameters. | ||
+ | |||
+ | 1. **Radius** | ||
+ | Expected diameter of dot to be detected in pixels. | ||
+ | |||
+ | 2. **CutOff** | ||
+ | Cutoff level for the none-particle discrimination criteria, a | ||
+ | value for each dot that is based on intensity moment order 0 and 2. | ||
+ | |||
+ | 3. **Percentile** | ||
+ | Larger the value, more particles with dark intensity will be | ||
+ | detected. Corresponds to the area proportion below intensity | ||
+ | histogram in the upper part of the histogram. | ||
+ | |||
+ | Try setting different numbers for these parameters and click | ||
+ | " | ||
+ | |||
+ | After some trials, set parameters to what you think is optimum. | ||
+ | |||
+ | ####Set Linking parameters | ||
+ | |||
+ | Two parameters for linking detected dots should be set. | ||
+ | |||
+ | 1. **Link Range** | ||
+ | | ||
+ | and reappears. If not, set the value to 1. | ||
+ | |||
+ | 2. **Displacement** | ||
+ | | ||
+ | to the next. Unit is in pixels. | ||
+ | |||
+ | After parameters are set, click " | ||
+ | |||
+ | ####Inspect the Tracking Results | ||
+ | |||
+ | When tracking is done, a new window titled " | ||
+ | bottom of the window, there are many buttons. Click " | ||
+ | trajectories", | ||
+ | trajectories will appear. | ||
+ | |||
+ | Select a region within the stack using rectangular ROI tool and then | ||
+ | click "Focus on Area". This will create another image stack, with | ||
+ | only that region. Since this image is zoomed, you could carefully | ||
+ | check if the tracking was successful or not. | ||
+ | |||
+ | If you think the tracking was not successful, then you should reset | ||
+ | all the parameters and do the tracking again. | ||
+ | |||
+ | ####Export the tracking results | ||
+ | |||
+ | To analyze the results in R, data should be saved as a file. To do | ||
+ | so, first click "All Trajectories to Table" | ||
+ | created. In this results table window, select `[File > Save As...]` | ||
+ | and save the file on your desktop. By default, file type extension | ||
+ | is " | ||
+ | "comma separated file", and this is more classic but general data | ||
+ | format which you could easily import in many software including R. | ||
+ | |||
+ | ####Further analysis | ||
+ | |||
+ | **OPTIONAL: | ||
+ | and a histogram of velocity distribution. | ||
+ | |||
+ | Particle Image Velocimetry (PIV) | ||
+ | ================================ | ||
+ | |||
+ | #### | ||
+ | |||
+ | There is a **PIV** plugin bundled with Fiji, but we use another PIV | ||
+ | plugin developed by **Qingzong Tseng**. You need several files for this | ||
+ | plugin to run. Here is the recipe for running the plugin in 64-bit | ||
+ | Windows OS. | ||
+ | |||
+ | 1. Go to | ||
+ | |||
+ | - < | ||
+ | |||
+ | and scroll the page down to “Installation” section. There you | ||
+ | will find links to **PIV\_.jar**, | ||
+ | file. Download these three files. | ||
+ | |||
+ | 2. Move or copy above three files to the **plugin folder** within | ||
+ | Fiji or ImageJ folder. | ||
+ | |||
+ | 3. Go to | ||
+ | |||
+ | - < | ||
+ | |||
+ | (or you could click “template matching and slice alignment” link | ||
+ | in the side bar in Zong’s page) and scroll down to | ||
+ | “Installation” section. There you will find a link to OpenCV | ||
+ | library for 64-bit windows. Download the file and unzip it. You | ||
+ | will find two files, cv100.dll and cxcore100.dll, | ||
+ | |||
+ | 4. Move or copy above two files, cv100.dll and cxcore100.dll, | ||
+ | the Fiji or ImageJ folder (**not the plugin folder**). | ||
+ | |||
+ | #### | ||
+ | |||
+ | 1. PIV plugin works only with two-frame stack. Reduce the stack | ||
+ | size by Duplicating only part of the stack. To do so: | ||
+ | |||
+ | - | ||
+ | duplication range only two frames e.g. 10-11. | ||
+ | |||
+ | 2. There are three different functions for doing PIV under | ||
+ | `[Plugins > PIV >]`, Cross-Correlation, | ||
+ | recommends not to use Cross-Correlation. We try to do PIV using | ||
+ | “Basic”. | ||
+ | |||
+ | 3. When you choose `[Plugins > PIV > Basic]`, you will see a panel | ||
+ | asking you to input parameters. There are three iterations, | ||
+ | first start to estimate displacement of signal with large grids | ||
+ | and then to finer grids. By starting from larger grids, | ||
+ | estimation procedure recognizes global movement first and then | ||
+ | more detailed movements in later iterations. Try to set windows | ||
+ | size so that it fits the size of the image. | ||
+ | |||
+ | 4. Final output is a text file. This could be saved anywhere, but | ||
+ | be sure that you know where it is. | ||
+ | |||
+ | 5. To plot the PIV field (Vectors), `[Plugins > PIV > Plot]`. Try | ||
+ | to plot without changing the default value. If the plot looks | ||
+ | bad with too large / small arrows, close the plot and do the | ||
+ | plotting again with sums adjusted parameters. | ||
+ | |||
+ | 6. **OPTIONAL** Since output is a text file of vector values, you | ||
+ | could import the text file from Excel or R to plot histogram of | ||
+ | velocities. | ||
+ | |||
+ | #References | ||
+ | |||
+ | Bise, Ryoma, Zhaozheng Yin, and Takeo Kanade. 2011. “Reliable cell | ||
+ | tracking by global data association.” In *2011 IEEE International | ||
+ | Symposium on Biomedical Imaging: From Nano to Macro*, 1004–1010. IEEE. | ||
+ | [http:// | ||
+ | |||
+ | Cheezum, M. K., W. F. Walker, and W. H. Guilford. 2001. “Quantitative | ||
+ | comparison of algorithms for tracking single fluorescent particles.” | ||
+ | *Biophysical journal* 81 (oct): 2378–88. | ||
+ | doi: | ||
+ | [http:// | ||
+ | |||
+ | Grabher, Clemens, Adam Cliffe, Kota Miura, Joel Hayflick, Rainer | ||
+ | Pepperkok, Pernille Rø rth, and Joachim Wittbrodt. 2007. “Birth and life | ||
+ | of tissue macrophages and their migration in embryogenesis and | ||
+ | inflammation in medaka.” *Journal of leukocyte biology* 81 (jan): | ||
+ | 263–71. doi: | ||
+ | [http:// | ||
+ | |||
+ | Hamilton, Nicholas. 2009. “Quantification and its applications in | ||
+ | fluorescent microscopy imaging.” *Traffic (Copenhagen, | ||
+ | (aug): 951–61. doi: | ||
+ | [http:// | ||
+ | |||
+ | Jaqaman, Khuloud, and Gaudenz Danuser. 2009. “Computational image | ||
+ | analysis of cellular dynamics: a case study based on particle tracking.” | ||
+ | *Cold Spring Harbor protocols* 2009 (dec): 65. doi: | ||
+ | [http:// | ||
+ | |||
+ | Keller, Philipp J., Annette D. Schmidt, Joachim Wittbrodt, and Ernst H. | ||
+ | K. Stelzer. 2008. “Reconstruction of zebrafish early embryonic | ||
+ | development by scanned light sheet microscopy.” *Science (New York, | ||
+ | N.Y.)* 322 (nov): 1065–9. doi: | ||
+ | [http:// | ||
+ | |||
+ | Meijering, Erik, Oleh Dzyubachyk, and Ihor Smal. 2012. *Methods for cell | ||
+ | and particle tracking.* Vol. 504. Elsevier Inc.. | ||
+ | [http:// | ||
+ | |||
+ | Saxton, M. J., and K. Jacobson. 1997. “Single-particle tracking: | ||
+ | applications to membrane dynamics.” *Annual review of biophysics and | ||
+ | biomolecular structure* 26 (jan): 373–99. | ||
+ | doi: | ||
+ | [http:// | ||
+ | |||
+ | Sbalzarini, I. F., and P. Koumoutsakos. 2005. “Feature point tracking | ||
+ | and trajectory analysis for video imaging in cell biology.” *Journal of | ||
+ | structural biology* 151 (aug): 182–95. doi: | ||
+ | [http:// | ||
+ | |||
+ | Sbalzarini, Ivo F. 2006. “A MATLAB toolbox for virus particle tracking | ||
+ | ICoS Technical Report Notice of Copyright.” *Technology*: | ||
+ | [https:// | ||
+ | |||
+ | Vallotton, P., a Ponti, C. M. Waterman-Storer, | ||
+ | Danuser. 2003. “Recovery, | ||
+ | tubulin polymer flow in live cells: a fluorescent speckle microscopy | ||
+ | study.” *Biophysical journal* 85 (aug): 1289–306. | ||
+ | doi: | ||
+ | [http:// | ||
+ | |||
+ | Vogt, Walther. 1925. “Gestaltungsanalyse am Amphibienkeim mit örtlicher | ||
+ | Vitalfärbung. Vorwort über Wege und Ziele.” *Wilhelm Roux’ Archiv für | ||
+ | Entwicklungsmechanik der Organismen* 106 (dec): 542–610. | ||
+ | doi: | ||
+ | [http:// | ||
+ | |||
+ | </ |
documents/121005advancedimg.txt · Last modified: 2016/05/24 12:46 by 127.0.0.1