User Manual for v1.0
Contents
1. Hold the Alt key when clicking the graph to create cursors direction Maximum bend direction 0 Finish Analysis Figure 15 Quantification of the maximum bend Left A diagram using the first bend as an example to show the definition of the maximum bend The red circles represent the first three markers along the spline Right Bend trace of the first bend of a wild type worm 25 sec Note that the bend trace is dominated by large amplitude alternating between approximately 50 to 50 degrees bends but accompanied by smaller and apparently random oscillations The x time and y angle values of the most positive and negative points of each dominant bend are displayed by mouse clicking at the bend trace Upon clicking the Finish Analysis button the difference between the averages of the most positive and negative values 94 61 degrees in this case is automatically determined 16 Worm Path The traveling path of a worm can be reconstructed using this function The plot may be based on the positions of either the centroid by entering O or any of the 13 markers along the worm spline Figure 16 The worm path plot does not indicate the starting and ending points of the worm movements which could be told from the txt file generated by clicking the Export Worm Path tab ure 2 VEN oss ure 3 alao z J Edit View Insert Tools Desktop Window Help a Edit View Inset Tools Desktop Window Help a d kh Aapa a2. is used in this context to select the centroid For example you could select the centroid head tip and tail by typing 0 1 13 into the box Calibration is entered simply by entering the x and y values separated by a space When all settings have been selected correctly simply choose the save path for the txt file and click Analyze and Export BP batchAnalyze_gui Batch Analysis 1 E 20130401 wt2_spline_orig txt a EA20130401 wt2 wt2 txt all 2 E 20130401 unc 9 2_spline_orig txt E 20130401 unc 9 2 unc 9 2 txt all m fileSelectWindow el X J Spline and Stage Selection Enter path to spline txt r Enter path to stage txt Browse a Add Recording B l Remove Recording f Edit Recording m Clear List Frame Selection Analyses V Avg Sum of Bends V Avg Amplitude V Frequency V Avg Speed J RMS V Total Distance V Maximum bend V Direction Metrics Options 15 Framerate 01 Point Number s 1 Bend Number s 3 6364 3 Calibration Save Path 3 recording 2 24 11 ZW WT analyses 4 19 2013 txt Analyze and Export Figure 21 Batch Analyze module Multiple recordings may be analyzed simultaneously using this module The user may select a list of recordings by clicking the Add Recording button in the main user interface eft pick the parameters for quantification and choose the Bend Number s for bend analyses and Point Number s for movement analy
3. means frames 31 to 80 Analytical features of this module are divided into two groups Bend Analysis and Movement Analysis n Bend Analysis the user may specify a bend number to analyze Any of the 11 bends formed by connecting the 13 markers along the spline may be selected Several functions may be performed with Bend Analysis including plotting the bend trace and bend frequency spectrum quantifying the magnitude and frequency of maximum bends quantifying the sum of all bends averaged over time and reporting bending activities as RMS and maximum bend for any selected bend 1 to 11 In Movement Analysis the user may reconstruct the worm travel path report the worm amplitude calculate the speed and distance of movements and calculate the duration and distance of forward and backward movements A corresponding stage file has to be selected to perform the movement analyses The stage file is automatically selected unless it is not in the default folder The various functionalities of the Analyze module are described below 13 Plot Bend Trace The bend trace is a plot of bend angle versus time for a selected bend Its appearance gives the user some rough idea about bending properties However its most important use is to serve as the basis for detailed quantifications of bending properties as described later A bend trace of the first bend also Known as head bend of a wild type worm is shown as an example Figure 13 File Edit V
4. EM 0 73333 6 1293 37 7883 29 9578 0 86541 30 7684 28 2214 4 0518 32 8417 39 3152 6 1755 24 0148 0 73333 2324 165 EN Steps F 42 74 14 0 8 6 8711 37 3702 34 5881 6 2977 26 4325 32 7555 10 431 27 4742 40 0386 15 1508 18 7634 0 8 2424 391 IEE Steps B 32 0 Sg 0 86667 4 4576 34 3038 39 3173 11 6873 18 6609 34 7807 16 6533 18 909 40 0308 23 6564 10 2485 0 86667 2498 763 EN Dist F 937 1386 5613 3756 M 0 93333 3 9973 31 6153 40 1114 20 7548 11 1509 33 3118 24 7283 6 5392 37 4752 32 7454 3 8218 2573 078 Dist B 365 0706 0 17 1 21 2736 14 1305 41 483 32 1934 2 5936 28 4208 33 2143 4 7802 26 7145 38 9749 17 6012 2660 864 i FA F 0 71965 1 ES 1 0667 23 3338 2 3167 39 2562 36 9654 9 838 18 9816 36 986 14 9514 21 1377 37 0054 27 4431 2747 646 EN B 0 28035 0 19 1 1333 19 543 11 0618 32 4884 40 136 18 604 12 4987 36 7416 21 4299 10 3656 37 8436 28 9547 19 CESME Sheet eee se eee wt 1 11 amp Figure 20 Data exported by the Analyze module The Analyze module may export three different data sets including Bend Data eft Worm Path middle and Results right by clicking the corresponding export tabs In the Bend Data spreadsheet the first column indicates time sec while the remaining columns show the angles of the 11 different bends at each time point In the Worm Path spreadsheet the first column indicates time sec while the remaining two columns indicating the x and y coordinates of the centroid at each time po
5. J 2 Search Documentation p G Reme O a coart Layout DEGA Help gt Request Support Gd ie D Ca Find Fies amp Ld New Variable Ley Analyze Code New New Open i jCompare impot Save E Racal eet EES FILE lt gt ja amp gt C gt Users Zhaowen gt Desktop gt Wormtracker2013 Current Folder Command Window _ Name Successfully produced spline z L a tet a analysis_gui fig j F analysis_gui m Processing frame 23 of 60 analysisLoop asv Lines split successfully analysisLoop m Produced centerline successfully fe ang m a Spline divided into segments arrow m F Successfully produced spline averageSumOfBends asv f averageSumOfBends m te biba Processing frame 24 of 60 batchAnalyze asv i ss FE batchAnalyze m Lines split successfully 4 batchAnalyze_guifig Progoceg Seen successfully fe batchAnalyze_gui m Spline divided into segments batchSpline_gui fig Successfully produced spline fe batchSpline_gui m batchTrack_gui fig batchTrack_gui m Processing frame 25 of 60 bendCursor_gui asv Lines split successfully bendCursor_gui fig Produced centerline successfully f bendCursor_gui m Spline divided into segments bkip2 2 fig Successfully produced spline bkip2 3 fig bkip2 4 fig __ blank_stage file txt Processing frame 26 of 60 Blister_GULfig z ig PRR Blister GULm nes split successfully eee b Produced
6. ag Heads SA09 4 8 08 af Worm path X marks starting point Worm path X marks starting point 1000 1400 900 1200 800 1000 700 800 600 600 500 400 300 200 1000 2000 3000 4000 5000 6000 7000 3000 4000 5000 6000 7000 8000 Figure 16 The travel path of a wild type worm A Travel path of the centroid path B Travel path of marker 1 The x and y axis labels are in micrometers The red x sign indicates the starting point of the worm path The movement analyses were performed at a frame rate of 3 Hz 17 Speed and Distance The Speed function measures the average speed of a worm in um sec The Distance function measures the total distance um traveled by the worm including both forward and backward movements as well as the net distance travelled the straight line distance from the first position to the last position of the worm over the analyzed period When accessing the Distance function through the Analyze not the Batch Analyze a plot of time versus absolute value speed will be produced Figure 17 T A Figure 1 o e xs File Edit View Insert Tools Desktop Window Help UGWMsSs ki ARVO d a E a g Worm speed absolute value 300 oO V wo a 2 5 3 Kia Time sec Figure 17 Plot of speed over time for an unc 9 fc16 mutant worm Direction Metrics The direction function measures the number of frames in which a worm is moving forward and
7. centerline successfully __ calibration _gui asv i i Spline divided into segments calibration gui fig A calibration_gui m Successfully produced spline Command History amfigure f plot xCenterLine return clc Figure 8 Matlab command line window Batch Analyze Worm Motion Worm Image Folder 120130401 wt Stage Data File E120120401wtt wtt bet FOR ANALYZING MAGES FOR REVEWING AND CORRECTING PREVIOUSLY CREATED SPLINES ie batch produced Spline Data Orig Fie 120130401 wtl bstchProduced_splineData_orig txt Spline Data Unsuffixed Fae E 2013040 wttibatchProduced_spineDeta bt Calibration um px 3 6364 3 6264 Frame Rate 15 Threshold swap 1 anuat _ swap 2 manua swap 3 kuanua swa 4 manus swa 6 anvai _ a aaaaa swap 6 Manual swap 7 kumana swap 8 kuanua swap 9 manual swap 10 manua Spine Output E20130401wt1_spine bet Browse Figure 9 Fit Spline module after spline fitting Each worm is superimposed by an x sign marking the head location and 13 markers representing the deduced spline 10 Figure 10 Correction of spline fitting A The snapshot of a frame from the Analyze module showing that the spline was poorly fit using a threshold value of 100 B A new window pops out upon clicking the manual button in A showing the original image C After changing the threshold fro
8. Track A Worm User Manual System Setup Track A Worm requires the following major hardware components A stereomicroscope A digital camera An x y motorized stage A PC computer installed with Windows 7 82 bit and Matlab including tool boxes for image processing image acquisition instrument control and signal processing The hardware used in our system is shown in Figure 1 For alternative hardware please consider choosing a fast PC with a large monitor e g 24 inch an x y stage with travel distances longer than the diameter of the petri dish used in worm tracking and a camera that has driver support for Matlab Compatibility with 64 bit versions of Matlab and Windows operating systems has not been tested Matlab drivers for the camera and stage are designed to be modular Therefore only a few functions must be replaced in Track A Worm software to accommodate different hardware Figure 1 Track A Worm hardware setup The major components include 1 a zoom stereo microscope SMZ 800 Nikon 2 a black and white CCD camera XCD V60 Sony 3 a x y motorized stage Optiscan ES111 Prior Scientific Rockland MA USA with a stage controller Optiscan II a universal specimen holder H473 and a stage stand H413 and 4 a PC computer not shown installed with Windows 7 32 bit and Matlab R2012b The stage has a travel range of 126 mm X 76 mm with a minimum step size of 1 um Software Installation Co
9. e data using the Fit Spline module by following these steps 1 select a worm image folder using the Browse bution on the top left which results in automatic entry of the corresponding stage data file into the Stage Data File box 2 click the Load button on the top right to observe the first 10 frames of the recording appearing on the screen without spline fitting 8 click the Load button in the section titted FOR REVIEWING AND CORRECTING PREVIOUSLY CREATED SPLINES i e batch produced to observe the spline fitting results displayed over the worm images The spline fitting files are automatically selected in this step although the user may override the automatic selection 4 inspect and correct the spline fitting results as described earlier and 5 save the verified spline data by clicking the Save Changes button at the bottom of the window Again the default folder and file name are entered automatically Note that the default folder for the spline files is now one level higher than the image folder which is the same as that for the spline files generated without using the Batch Spline module In other words no matter Fit Spline or Batch Spline was used for the initial automatic spline fitting the final spine data are saved in the same folder and in the same format 12 Analyze The Analyze module Figure 12 allows the user to perform all the analyses built into Track A Worm It is typically used to perform quick analyses for single record
10. iew _ Tools Desktop Window Help OGHS k ARC98L4 a 08 a0 Bend Trace 80 60 40 o D D q D m gt D isa 10 15 Time sec Figure 13 Bend trace of the first bend of a wild type worm 14 Dominant Frequency A worm typically bends with one large dominant motion with smaller random movements in between Figure 13 Bend Frequency reports the frequency of the dominant bends in a bend trace For example a plot of the frequency spectrum Figure 14 for the bend trace shown in Figure 13 indicates that the dominant bend frequency is approximately 0 4 Hz File Edit View Insert Tools Desktop Window Help CODEER F FALIE 4000 3500 3000 2500 lt 2000 v gz Z gt 2 1500 Frequency Hz Figure 14 Bend frequency spectrum of the first bend of a wild type worm This graph is generated by Fourier transformation of the bend trace shown in Figure 13 The Fourier transformation always produces a mirror image Thus the user should disregard the second half of the graph The main peak of this graph indicates that the dominant bending frequency is 0 4 Hz Root Mean Square RMS The bend frequency described above ignores the smaller and apparently random bending activities in the bend trace However changes of the smaller bends might be characteristic of some mutants Therefore it is important to include them in the quantification Track A Worm uses RMS to meas
11. ings or to perform functions unique to this module including plotting bend trace bend frequency trace and worm travel path Another module known as Batch Analyze described later is used for simultaneous analyses of multiple worms Analysis General Settings Spline Path E 20130401 wt2_spline_orig txt Frame Rate 15 Note Choose the _orig export file not the unsuffixed standard file all Frames to Analyze Bend Analysis Movement Analysis Stage Path E 20130401 wt2 wt2 txt Bend Number 1 11 1 Spline Point 1 13 or 0 for centroid 0 X 3 6364 Y 3 6364 Piot Bend Trace Sum of All Bends Dominant Frequency and RMS Maximum Bend Plot Worm Path Worm Amplitude Export Bend Data Speed and Distance aa Export Worm Path Messages Messages __________ Freq 0 43 Hz RMS 26 77 deg Forward 74 5613 38 um Backward 0 0 um Export Results Export File Enter Path to Export txt Comments Manually Produce Maximum Bend During Export Save and Append Figure 12 Analyze module user interface The user first needs to select a spline file enter the frame rate used for the recording and select the range of frames for analysis under General Settings As indicated in the General Settings box spline files with the _orig suffix should be used The frame range to analyze may be entered as either all or two numbers separated by a space e g 31 80 which
12. int The Results file could contain data from multiple recordings show are data from two recordings In the Results spreadsheet the first column shows various row labels including the point number 0 or centroid in this case the bend number bend 1 in this case comments anything from the user recording length sec average sum of all bends degrees frequency of dominant bends Hz mean amplitude of maximum bends degrees average amplitude um ratio of amplitude over worm length A L average speed um sec total distance traveled um the number of forward steps the number of backward steps the total distance of forward movement and the total distance of backward movement The remaining two columns show the data from one wild type worm and one unc 9 f 16 mutant worm 20 Batch Analysis The Batch Analyze module Figure 21 may perform all the quantifications described for the Analyze module It is better suited to analyzing large sets of data Recordings are added via a small pop up interface Figure 21 Analyses are chosen via checkbox Bend Numbers are chosen using standard Matlab syntax You can choose individual bends using the format 1 3 5 indicating discrete points You can choose entire ranges of bends using the format 1 5 indicating a series of connected points Remember that there are a total of 11 bends for a worm spline Point numbers are also chosen using the same syntax However the choice 0
13. interface typically contains several fields displaying the path and name for all the related files However only the path and name of the base file have to be entered manually Upon pressing Enter the path and name of the related files are automatically entered or updated with the updated fields flashing green momentarily This feature is included to save time and prevent user errors To run Track A Worm click the desktop icon that you created earlier The first window that appears on the screen Figure 2 allows the user to launch various modules from it including Calibrate Record Playback Batch Spline Fit spline Analyze and Batch Analyze The procedures for using each module are described below Worm Track Launcher Determine the correct micrometer pixel value for proper camera control and tracking Interface with camera and stage to record worm movement Playback and manipulate recorded data Find optimum threshold and video segment for spline fitting Select a group of video recordings to generate splines which may be reviewed corrected at a later time Track worm by fitting and reviewing splines Export spline data for analysis Analyze worm movement including angle direction frequency and other features Analyze worm movement in large batches using a convenient GUI Figure 2 Track A Worm launcher Calibration Prior to making a recording the correct conversion factor from pixels to micrometers need
14. ithin a couple of seconds immediately after clicking the Start button but before the tracking has actually started to ensure the worm is still in the camera field However as long as the worm has a small part in the imaging field at the moment when the camera becomes ready it can be identified reliably and moved to the center of the field automatically A faster starting camera would improve this aspect of the user experience After the designated recording period is over the Record module will save the files The Status Panel will indicate when the process is completed To begin a new recording you must click Connect Camera and update the base file name e g change wt1 to wt2 and if necessary change the folder name Canceling a recording It is sometimes necessary to terminate a recording before it ends To do so click on the MATLAB command window to make it active and then press CTRL C A series of error messages will appear and the recording will be canceled You will then need to click Disconnect Camera and Connect Camera This returns you to Step 3 Playback The Playback module Figure 5 allows the user to load a recording for playback evaluate the threshold for binary conversion specify a frame or frames to display and cut out undesired frames The user interface of this module includes three sections Video viewer Playback Control and Frame Selection The video viewer can display images in either original or binar
15. l all Els 0 066667 26 4087 25 9106 41 9361 13 2893 14 2101 38 2189 25 3588 0 75895 37 6523 39 2932 1 3161 0 066667 1491 533 Ely Rec Length 25 25 EM 0 13333 25 7933 6 3766 41 5066 25 0605 3 0771 36 0512 32 5234 7 6989 24 8863 44 7558 10 8361 0 13333 1567 374 EH Avg Sum of Bends 361 8107 291 5237 5 0 2 14 1523 3 0087 34 1651 33 6641 4 2057 32 0823 35 328 15 7422 22 0721 40 5878 27 6807 0 2 1647 369 il Freq B1 0 56769 0 42576 Sim 0 26667 23 8387 7 6838 28 4734 34 7374 12 8273 24 0094 37 7231 24 4634 11 9002 41 0159 32 0724 0 26667 1722 677 a RMS B1 38 0552 26 771 vam 0 33333 28 5681 13 2812 19 6454 35 4827 20 938 16 9324 38 002 29 3519 7 0656 34 7338 35 5206 0 33333 1796 074 Max Bend B1 103 8606 94 6075 8 0 4 45 1929 20 0146 11 3059 33 8503 28 0551 7 0528 35 4004 32 8593 5 5649 28 5367 40 6183 8 0 4 1891 292 Ea Amp 420 3737 222 2673 Si 0 46667 58 1149 32 0104 2 9473 32 4019 29 4031 4 6489 30 2963 36 3047 15 8147 22 5036 40 0711 0 46667 1973 898 EB A L 0 46646 0 22854 fm 0 53333 29 9535 44 8963 9 4022 26 0542 34 098 11 2649 19 6694 40 678 22 3462 14 7331 37 4817 0 53333 2046 729 N Speed PO 52 7923 227 5693 11 0 6 1 8099 45 4493 21 0938 16 4584 36 3154 18 5088 11 7358 39 8616 29 6809 10 887 35 2784 0 6 2151 629 ISE Net distance traveled 496 6732 5314 7515 em 0 66667 5 4324 38 7754 27 761 10 1044 33 1234 24 0274 2 8434 38 6403 33 8859 1 8781 32 3265 0 66667 2236 214 bea Dist PO 1302 2092 5613 3756
16. m 100 to 70 in the open box and clicking the Adjust button 13 spline markers are placed appropriately over the worm with the head indicated by an x If both the spline fitting and head assignment were satisfactory click the Keep button in the small window If the spline fitting was good but the head assignment was wrong click the Swap button If the spline was not successfully fit try a new threshold or click the Start button to do manual spline fit You will need to close and reopen the window shown in B to redo the analysis D Upon clicking the Start button in B the user may provide guiding markers as shown by clicking along the midline of the worm 9 to 15 times gt 15 is fine in the head to tail direction Upon clicking the Done button the head will be correctly identified and 13 markers will be placed appropriately along the worm This approach is more user dependent but always works 11 Batch Spline The spline is fit through complicated computational processes Automatic computation is the most time consuming step in obtaining satisfactory spline fitting In the Fit Spline module described above the user needs to select the image folder and inspect correct the fitting results for one recording before moving onto the next one Therefore frequent user attention is needed to upload new recordings The Batch Spline module Figure 11 allows automatic spline fitting for many recordings in a batch The user can then choose a convenie
17. me number the open box and the total number of frames are displayed below the Load button The frame for display can be selected by typing into the open box and clicking Apply The Threshold box allows the user to enter a threshold value for binary conversion Effective threshold values vary depending on lighting conditions A higher threshold should be chosen for brighter images and vice versa If there is a strong contrast between the worm and background then a wide range of thresholds may work Conversely only a narrow range of thresholds may work If the brightness level is set too low portions of the worm may blend into the background resulting in an incomplete silhouette Figure 5 middle If the brightness level is set too high shadow around the worm may be included as part of the worm resulting in a wider worm body in the binary image The worm in the binary image should appear in solid black color and have a shape closely matching that of the original worm Figure 5 right Track A Worm has built in mechanisms to automatically adjust contrast eliminate background noise and fill in holes in the worm body so that a well represented worm profile can generally be produced The small check box before the Use Thresholded Image determines whether the original or binary images are displayed in the Video Viewer The small check box before Show spline if available determines whether the 13 markers representing the spline are overlaid on the wo
18. nt time to inspect and verify the results El betchSpline_gui af L3 7 Batch Spline EA20130401 wtt T 090 FR 15 E120130401 wt2 T 100 FR 15 Dic i H Folder Select Window EA 20130401 wt3 Frame Rate 15 Threshok 100 3 6364 3 6264 Calibration Done asd remove Analyze Figure 11 Batch Spline module This module is used to perform automatic spline fitting for multiple recordings in a batch The main user interface for this module is shown on the left Clicking the Add button opens a new window right in which the user may select an image folder enter the frame rate used for recording the images and choose a threshold for binarization for a specific recording Upon clicking the Done button the selected image folder with the frame rate and threshold is added to the open box of the main user interface Many recordings may be added by repeating these steps If an undesired image folder was selected by accident it can be removed by clicking the Remove button and entering a number matching its position in the list Upon clicking the Analyze button Track A Worm starts automatic spline fitting The results from the Batch Spline are saved in the original image folder as two files batchProduced_splineData txt and batchProduced_splineData_orig txt which are with and without stage compensation calibration respectively The next step is to inspect and correct the automatically generated splin
19. py Track A Worm program files to a directory of your choice Create a shortcut to Matlab exe typically in C Program Files Matlab bin Matlab exe on your desktop Right click the newly created icon and modify the field Start In to reflect the directory where Track A Worm is located Track A Worm can now be launched by double clicking the shortcut icon We suggest that you rename this icon to Track A Worm Operating Instructions File structure Track A Worm saves files to and looks for them in a predetermined hierarchy Before the recording you need to manually create a main folder for saving your data such as 20130401 based on the date of the recordings Suppose that you would like to use wt1 as the base name for the first recording of wild type worms you simply need to type wt1 after the folder name e g 20130401 wt1 into a specific open box in the Record module before the recording details in a later section A subfolder named wt1 would be created automatically when the recording is started This subfolder contains 1 recorded image files that were automatically named as img001 bmp img002 omp img003 bmp etc 2 a subfolder named removed containing removed files This folder is automatically generated when the Cut function in Playback module is used to remove some files 3 an active stage file named as wt1 txt and another stage file named as wt1_orig txt which is the original
20. rm When selecting frames using the Frame Selection section it is important to remember that stage movement data are tied to the image sequence Therefore to cut a specific portion of the recording for subsequent analyses the stage movement file must be updated When entering the frame numbers for the lower and upper cut edges the number for the lower cut edge should be fn 1 whereas that for the upper cut edge fn where fis the frame rate and nis an integer Therefore if the frame rate is 15 and you want to keep the data from frame 20 to 80 you should actually select from frame 31 to 75 Otherwise stage movement timing becomes disrupted The action of cutting creates a subfolder within the data folder named as removed The updated stage file reflecting the selected data is saved using the Original file name whereas the original stage file is renamed by appending orig to the file name Therefore subsequent analyses should be carried out on the unsuffixed stage file The orig stage file and the contents of the removed folder can be used to manually restore the data folder back to its original state Note that the path and stage file name is automatically entered in the Frame Selection section although a Browse button is available to override the default choice Fit Spline The Fit Spline module Figure 6 takes the recorded images and the stage file as input to produce two spline files Both spline files display x y po
21. s to be determined using the Calibrate module Figure 3 The Calibrate module receives input from the camera Operation is as follows 1 View iY D 7 8 Place the stage under the microscope objective and align its x axis straight relative to the camera lt is important to maintain the stage in the same orientation during subsequent image recordings although the exact location of the stage does not matter Click Connect Camera and wait for the camera to start up Place a stage micrometer under the microscope and align it in either the x or y direction Adjust microscope magnification to a desired level Click first on one end of the micrometer and then on the other end to observe a straight red line drawn between them Enter the actual distance spanned by the drawn line into the x or y box under Calibration um You may leave the other box blank e g leaving the y box blank when calibrating for x Click Done to observe the calibration displayed in the Results box Repeat the procedures to perform calibration for the other direction which is necessary Click Disconnect Camera before exiting the module Write down the calibration results so that you may enter them into the other modules at a later time You do not need to recalibrate if you are going to use the same microscope magnification settings in subsequent experiments Calibrate Figure 3 Calibrate module This module i
22. s used to determine the correct conversion factor from pixels to micrometers for the x axis and y axis Record The Record module Figure 4 receives worm position information from the camera sends commands to re center the stage at 1 sec intervals and outputs sequential images to a user designated folder in BMP format It also saves a text file containing all the stage movement information Camera Controls Brightness 1200 Exposure 12 Framesisec 15 a m Record Time sec Select Bmp export folder E 1 20130401 wt3 Select export for stage movement E 20130401 wt3 wt3 txt Calibration um px x 3 6364 Y 3 6364 i 2 Start ma Tracking and Acquiring Figure 4 Record module This module is used to record worm images To add a new recording e g wt3 to the same folder e g 20130401 after finishing one recording e g wt2 the user only need to change 2 to 3 in the open box before the first Browse button and press the Enter key to observe that the stage movement file name is updated automatically A subfolder named as wt3 is automatically generated at the start of the recording A typical workflow is as follows 1 Click Connect Camera and wait a few seconds for the Camera indicator in the top right corner to change from red to green 2 Click Connect Stage and wait a few seconds for the Stage indicator in the top right corner to change from red to green Click Center to ens
23. sed for the recording and choose an appropriate threshold for binarization The threshold value could be evaluated with the Playback module Track A Worm has an internal mechanism to vary the threshold by as much as 20 from the initial guess if a valid spline cannot be found using the selected threshold Therefore a spline can generally be fit even if the initial threshold was not optimal although the time to fit could be significantly longer due to repeated attempts Since lighting conditions were generally similar for all the experiments of a batch one threshold value may suits all the experiments Press Start Analysis The initial detection window will appear with 13 markers along the spline and the candidate head indicated by an x Figure 7 You are prompted to either keep the head assignment or swap it with the tail Since the head assignment in a preceding frame also serves as a reference for that in the following frame it is important to verify the head assignment at the initial detection window ta Figure 1 File Edit View Insert Tools Desktop Window Help DOGaS k ASACU90R4 2 08 ao Figure 7 The initial detection window for verifying head assignment This window appears at the beginning of spline fitting Click Keep if the head assignment was correct but Swap if the head assignment was wrong This window then disappears which is followed by automatic spline fitting for the remaining frames After clo
24. ses The small window right pops out upon clicking the Add Recording button in the main user interface to allow the user select specific spline files suffixed with orig When a specific spline file is selected the corresponding stage file is automatically selected Frame selection is performed using the same syntax described earlier e g enter 1 50 to select frames 1 50 or all to select the entire recording 21
25. sing the initial detection window the software will run on its own for a period of time Each frame typically takes 1 second on a Core i7 920 PC but longer if the image is more challenging During this period of time the Fit Spline user interface appears to be stationary The progress of spline fitting is displayed real time in the Matlab command line window Figure 8 When the analysis is finished a spline is displayed over each worm image with the head marked by an x sign Figure 9 The user should then inspect the spline fitting and head assignment results for all the frames of this recording Five frames are advanced each time the Down button is clicked Figure 9 Potential errors can be corrected as follows Incorrect head assignment Click Swap to fix it This action also causes head and tail swap for all the subsequent frames since head assignment in each subsequent frame was partially based on head position in the preceding frame during the automatic analysis Failed or poor spline fitting Spline fitting for a specific frame may be fixed by first clicking the Manual button below it and then following the instructions in the legend of Figure 10 7 After you have inspected spline fitting for all of the frames click Save Changes to save the spline file Note that the default folder and file name are automatically selected by Track A Worm although you may override it which we do not recommend MATLAB R2012b H
26. sitions of the 13 spline markers in a spreadsheet format One file e g wt 1_spline txt contains stage compensated spline data whereas the other suffixed with orig e g wt 1_spline_orig txt contains the raw spline data without stage compensation and calibration Batch Analyze Worm Motion Worm image Folder EA20130401 wt1 C Browse Stage Data File EA20130401wt1 wt1 txt C Browse FOR ANALYZING MAGES FOR REVIEWING AND CORRECTING PREVIOUSLY CREATED SPLINES i e batch produced Spline Data Orig Fite E 120130401 wt1 batchProduced_splineData_orig txt Spine Data Unsuffxed File E120130401 wtl batchProduced_spiineData txt Calibration umpx 3 6364 3 6364 Frame Rate 15 Threshold 100 Start Analysis Figure 6 Fit Spline module prior to spline fitting The Fit Spline module may perform two functions 1 generate splines for individual experiments and 2 verify and correct splines generated by either the Fit Spline or Batch Spline module In this section we describe how to use Fit Spline to generate inspect and correct splines for individual experiments A typical workflow is as follows 1 Select a specific folder containing worm images The corresponding stage file is automatically selected 2 Press Load to observe the first 10 frames of the specific image folder appearing on the screen 3 Enter the x and y calibration information and the frame rate u
27. stage file renamed after performing the Cut function in the Playback module The stage files contain stage movement information and 4 two spline files generated using the Batch Spline module including batchProduced_splineData txt and batchProduced_splineData_orig txt Upon finishing spline inspection and correction using the Fit Spline module two spline data files are automatically generated and saved in the 20130401 folder as wt1_spline txt and wt1_spline_orig txt which contain the x and y coordinates in micrometers of the spline over time with and without stage movement and calibration compensations respectively In other words the unsuffixed file is both stage compensated and calibrated while the orig file is in raw pixels only These two files are saved in this folder so that the user may find spline files of all the recordings of related experiments in the same place The two splines generated using the Batch Spline module are saved in the wt1 subfolder instead of the 20130401 folder because the spline fitting and head assignment need to be verified by the user at first To maintain the integrity of the folder and file structures the user should avoid changing either the folder hierarchy or folder file names using Windows Explorer When it is necessary to do so be sure to make corresponding changes to all the related folders and files to reflect the convention described above A graphical user
28. the number of frames it is moving backward Directionality is determined by the velocity vector a straight line from the centroid of the current frame to that of the next frame and the head vector a straight line from the current centroid to the current head tip position The worm is considered to be moving forward if the projection of the head vector onto the velocity vector is positive and vice versa Figure 18 The direction function also outputs the total distances of forward and backward motion 18 Velocity vector Figure 18 Determination of movement directions The red circles represent the positions of the centroid at the current frame 1 and the next frame 2 The green circle labeled a and the blue circle labeled b show that the head vector is projected to the positive and negative sides of the velocity vector respectively Worm Amplitude Amplitude is another measure of worm curvature A highly curved worm tends to have large amplitude and vice versa Amplitude is determined by first finding the velocity vector as described above Then a box is drawn in parallel with the velocity vector to enclose the widest points of the spline not the actual worm body Figure 19 The width of the box is the worm amplitude Because the amplitude could be affected by the length of the worm Track A Worm also reports the ratio of the amplitude over the length of the worm A L Velocity Vector N A Amplitude Fig
29. ure 19 Determination of worm amplitude 19 Results display and Export Either a plot or the calculated result is displayed instantly upon clicking a specific functional tab in the Analyze module A plot is displayed in a new window whereas calculated results are shown inside the Messages boxes The Analyze module may also export the data for further analyses by clicking three different export buttons The Export Bend Data button is used to export the angle values over time for any of the 11 bends The Export Worm Path button is used to export the x y coordinates over time for the active Worm Path plot A worm path plot must first be produced to use this function The Export Results button is used to export all the remaining results not included in the other two export functions When using the Export Results function the user may add comments choose to include manually produced Maximum Bend and append results from multiple experiments into the same file The user should not modify the export file manually using Excel or any other application before the completion of appending new data to prevent corruption of the formatting All the export files are tab delimited txt files which may be dragged into Excel to open Figure 20 B re D 3 F G H j J K L A Cc Time 1 Bend 1 2 3 4 5 6 7 8 9 10 11 a Name unc 9 2_spline_orig wt2_spline_orig 2 O 1 1357 40 1985 34 7774 5 987 23 9398 37 7382 18 2668 5 7555 40 4914 37 0642 6 5829 ya Comments al
30. ure all bending motions regardless of their amplitudes using the following formula n y2 n where n is the total number of points and refers to each individual point While RMS is potentially more sensitive in detecting phenotypes of many mutants it has limitations in distinguishing between high frequency low amplitude and low frequency high amplitude behavior 15 Sum of All Bends The Sum of All Bends is a useful metric for quantifying the bending behavior of an entire worm It is simply the sum of all the 11 bends of a worm averaged over time The Sum of All Bends metric uses the absolute values of the bends regardless of the directions of the bends Maximum Bend It is often intuitive and useful to quantify the maximum sweep of a bend Clicking the Maximum Bend button opens a bend trace of the selected bend which allows the user to place markers at the most positive and negative points of each dominant bend by mouse clicking while holding down the ALT key Figure 15 Maximum Bend is calculated as the difference between the averages of the positive and negative values This measure is only useful in worms that exhibit organized bending behavior If the behavior is very chaotic without obvious extrema of movement the RMS measurement may be more useful Before quantifying the maximum bend it is important to define a threshold For example the user may choose the most positive and negative peaks larger than 20 degrees alternatively
31. ure that the stage is at its central position Enter the x and y calibration values found in the Calibrate module Adjust the relative brightness level exposure frame rate and recording time settings as needed You must click Apply to make the new settings to take effect The brightness should be set to such ae ae 5 a level that the worm appears as a dark grey color with good solidity as shown in Figure 4 The frame rates that may be chosen are 1 3 5 and 15 When choosing a frame rate the user should consider about potential stage limitations Some stages may be unable to initiate and complete a movement at a sufficiently fast speed and thus resulting in image blurring due to stage movements There is no such problem with the Prior stage shown in Figure 1 6 Select the BMP export folder and enter the base file name for the recording The corresponding folder names in the Select export for stage movement are entered automatically 7 Manually by hand not using the Align Stage buttons move the stage to find the worm Remember to maintain stage orientation as you move it When the worm is within view click the Start button You may observe automatic adjustment of the exposure conditions for the first couple of frames The Sony XCD V60 camera takes a few seconds to begin recording after the Start button has been clicked During this delay period the worm may move away from the center of the imaging field You may manually move the stage w
32. y format To use this module the user first needs to load a recording using the Browse button under Playback Control Playback Playback Playback IN Playback Control Playback Control Playback Control Frame Pause 15 Pisy Frame Frame Pause 15 Pisy Frame Frame Pause 15 l Play Frame E120130401 wtt Browse E120130 01 wtt Browse E120120601 wtt Browse Loss 5 45 ETADON Threshoki 80 L Use Thresholded Image Threshokt 0 4 Use Threshokled image Threshold 70 J Use Threshoiied image Show splines if available __ Show spines if available Show spines if available Frame Selection Frame Selection Frame Selection 12013040 t wttiwtt bet EA20130401 wttlwtt txt Browse EA20130401 wttlwtt bet 1 1 cut 1 focus 1 1 cut Reduce Frame Rate Reduce Frame Rate Reduce Frame Rate Figure 5 Playback module Left An original worm image is displayed Middle The threshold was set too low resulting in an incomplete worm silhouette in the binary image Right The entire worm appeared as solid black after the threshold was increased The open box between the Play and Pause buttons is for choosing a relative playback frame rate The actual playback frame rate also depends on the speed of the computer The Frame and Frame buttons are used to move forward and backward a single frame The current fra
Download Pdf Manuals
Related Search