Manual ADT3D
Contents
1. BH Difference Vector Space Viewer pce ee File View Tools Units a 7 5 ra In the View menu of the Difference Vector Space Viewer the user can select different visualization options Cluster points option colour all difference vectors assigned to found clusters Activation of Noise points visualize all difference vectors grey points which were not assigned to any cluster and represent noise in the data Cluster option shows centres of found clusters as small red spheres Additional cluster option visualizes extra clusters created by the user at the beginning there are only automatically found clusters therefore no additional clusters exist In the Difference Vector Space Viewer the volume can be rotated using left mouse button Pressing Escape button on the keyboard changes the visualisation mode into editing mode In the editing mode the clusters can be added deleted etc Whenever there is a need to rotate the volume by pressing the Esc button the mode can be switch back to rotation The mode can also be changed by pressing the view gt toggle mode button A certain cluster can be selected a 3D box is placed around it by clicking on it It is easier to handle the clusters when only cluster points and noise points are blacked out Keeping Shift button of the keyboard pressed several clusters can be selected When two clusters are 36 eT selected an additional text box appears in the menu bar on the right side showing th2. 68 72 76 80 84 88 90 92 96 98 128 130 132 134 136 140 144 148 47 cell angles in degrees mapping colums rows sections on axis x 1 y 2 z 3 space group number 0 for images number of bytes used for storing symmetry operators number of bytes in extended header This is an important number It defines the offset to the first image creator id extra 30 bytes data not used number of bytes per section in extended header number of floats per section in extended header extra 28 bytes data 47 4 byte floating point 4 byte floating point 2 byte integer 2 byte integer 2 byte integer 2 byte integer 2 byte integer 2 byte integer 4 byte floating point 4 byte floating point 4 byte floating point 4 byte floating point 4 byte integer char min4 max4 idtype lens nd nd2 vd1 vd2 tiltangles 9 ZOrg xorg yorg nlabl data 10 80 152 156 160 162 164 166 168 170 172 208 212 216 220 224 48 For Truelmage the identification is 257 divide by 100 to get float value used to rotate model to match rotated image origin of image used to auto translate model to match a new image that has been translated number of text labels with useful data 0 10 10 text labels with 80 characters The extended header Total number of extended headers one for each image is 1024 Please note Some values have rather odd numbering That is becaus
3. Data reduction Ctrl O Ctrl S Import gt Export r e Open CELL file Save CELL file Open HKL file Save standard HKL file Save extended HKL file Ctrl W Alt W Close Current Close Current Without Saving Test 3D View Exit Ctrl Q Intensity data is written out in a standard crystallographic 314 2f8 2 format When the data set contains high intensity values which go over 8 digits the program automatically changes format into 314 2f8 1 format to avoid clipping of strong reflections When further space is needed to write out the intensities the format is automatically changed to 314 2f8 0 Further reduction can be done by division by 10 100 etc If a format has to be changed a message appears in the log panel Below an example of hkl file is shown This filed can be used as input for crystallographic structure solution programs like SIR provided the atomic scattering factors for electrons are available 43 An extended hkl file does not follow the additional information about the reflections z kd gt lt BaSO4 PED_sh_rot45_bi_re hkl WordPad B Start Ansicht g Courier New ju Al a EEE d Suchen ab Ersetzen Sac Einf A n Ar z Einf gen akad FK abe oe Er S ET AASA ii alle ausw hlen Zwischenablage Schriftart Absatz Bearbeiten OE coe sa as le ne FAES MA a A E LES A GE GOE S B 10 S a ait cil sive Se HES 15 VD ee 3 3 12059 143 3 F 3 2 7867 7 88 8 8 0
4. User manual version 1 3 2 11 January 2012 Table of contents Ie TMtROGUCHIOM rrrrerrdenter er E ee denderendaddeiek A A 3 Dy AIMS CAM osnan eeen ebde seta ade handed aad arrie ie 5 AMEE O A Aa E E A E E E EA T 5 22 ADT D insta liio Peri a e Borsa cs a e a a siebe 5 JA lew words about the ADT experiment ar ureters a a 6 de MZ IVAN C dati CONVE REN e E E E E 7 5 Starting ADT3D the program philosophy recisrsisyrint a a n eerden aan 10 OA DONEN EE 12 Le 2D MUMCUONS aan a E eedt deense ETR 13 FM AVA OUD zeten reeden diketon dheindesbeied eten A 13 Ted AUC ICOM be ING aa e E euledeieluan ian sassumauidoniian cues emwidentianeses wauidenvinnaes 14 7 3 Tilt axis determination rotation to the tilt axis oan vennnssennserennerennen 16 7 4 Reconstruction of the volume data aaneen eeennssennserennseennnerennerennen 18 eS 9D WEB renane e a E a E E a 19 TD T O eren e aio dct crete E A a E a r A eee ten eRe 25 Peak Se AL Cl le eassa a a A 29 9 2 Manual assessment of the unit cell vectors aans oeens eenn erennssennnerennenennen 28 9 3 A tomate umt Gel GETS PINT ALI OM rnaar ended ieee dasa neee ee dan dna eea doands 29 Ode Example CIUSter ns Parameter TUNNE onenen na e E E ET E 32 9 5 Clustering parameter tuning statistical approach sseesssesseesssersserssersserssersserssees 34 Osi ference VECLOL SPACS VIE WEE naarotanmatenn iaaa aa a 36 ht Eranstormmdtom ofthe Umit Gel mensne eh ekain 39 2 Extraction Or HKL In
5. 24 View The View option allows toggling between the orthographic and perspective projection of the diffraction volume er RS a io RE gt 3 a N Se SA NJ ze 2 N R 4 4 wR QE g EET EE A TO eine neo eee en ima h OA DO Bt a gt rn o Pewee a amp p ee meren ae a ee PEPETA werd bg gt P fer k i Pa l EN Ald eN ne a s CE a RO ver Bn RA te anas e a bi h a S hn nt aA n b RM t t amp sa r S a t T x i w 8 vps 4 24 25 9 3D functions When a volume MRC file is created or opened a selection of 3D functions peak search cell determination and peak integration become active These routines should be performed in the presented order one after the other 9 1 Peak search The automatic cell search procedure starts with a search of peaks in 3 dimensions The function can be called by pressing the Find peaks icon or from the menu pressing the 3D functions gt Find Reflections BY Find Reflections options Units pixel voxel angstrom Cut off v cut off zero beam Distance 26 limit resolution Distance 425 Find reflections Min volume 5 Max volume 200 Threshold 500 Reset Cancel Found reflections dropdown menu of the 3D viewer correlates with the Find peaks function of the main window menu and Options gt Find reflection parameters panel The parameters in the reflection parameters panel ca
6. Camera length in mm e High tension kV Diffraction lens value in if the used is intended to work with the fine calibration Otherwise this parameter is irrelevant MN Tilt series INFO staring tit angle in deg NEGATIVE POSITIVE Tit step 1 ema Tilt seri Pixel size in 1 nm 0 025 Camera length in mm 50 EN Tilt seri High tension kW oo Diffraction lens value in Yo 39 Converter INFO Stack of TIF files with a generc name MAAD T3DAD ata CustomersPica_Maurosht datasMDT_ ie converted into MAAD TSDAD ata Customer Fisa Mauros datasMDT mre If necessary please ask the developers to add additional options to wire other parameters into the MCR header The present version of converter is limited to 500 patterns per tilt series 10 5 Starting ADT3D the program philosophy Currently only MRC files with the format declared in the Appendix are read and processed by ADT3D MRC diffraction data stacks can be created from TIFF files using the converter supplied with the ADT 3D package Two types of MRC files are used by the program Stack data tilt series and Volume data Stack data A file which contains a series of electron diffraction patterns images acquired during the tomographic tilting of a crystal The x and y dimensions are the size of the image in pixels all diffraction patterns within a tilt series must have the same number of pixels and z comprises the total number of
7. Ze 176 6 8 0 4 58521 4 241 4 8 0 6 90034 4 300 0 8 0 6672 2 81 1 8 2 853272 92 2 8 5 1 32939222 a Re ja See 8 6 0 75490 0 274 4 2 1 33340 0 182 2 7 2 2143638 8 H Gs eee F 2 3 61110 0 247 7 F 2 4171477 7 414 4 J 2 6 5168 8 ee Se E F 3 2 48922 2 aat X E Einf gen Blo s Start Ansicht Courier New IE gt BaSO4 PED sh rot45 bi re extended hkl WordPad Al aA F K U be Lr Ar tz G4 Suchen ab Sac Ersetzen mi KAI mi I il ih Il 4 m 3 4c wo fb 5 Zwischenablage Schriftart Absatz Bearbeiten Extended ADT3D HKL File Version 1 0 5 A RY SY SEY t It i J2 i13 14 A h k 1 intensity sigma intensity original center of integration box x y refined center of integration box xly tilt angle size of integration box x y eliptic integration area x y angle 322720539 145 ATA Sst SE ees KE 9 4 2 7867 88 66 ae tar ar seer EE oo 8 0 2 31085 176909 E ka Re se Ee X 8 004 S851 241 89 xix se Ke Se KOSS x B 0 6 900342300 057 x SE KME t 8 A D 6672 GL 6023 KEK SS ee KX B A S532 AAO ate MEEK See HK 8 5129392 171 441 Sees Kee KE 8 6 0 75490 274 754 x ee Ee KE KX 7 2 1 33340 182 592 x x xxxxxxxxxx 44 The user is generally free to select any cell defined by reciprocal basis vectors but for crystallographic computations it is necessary that
8. by clicking on the run centering button in the top right corner After the patterns are centered the original diffraction patterns are placed on the left side and the shifted patterns are on the right At the bottom of the window two bar plots appear the shift vector x and y coordinates The user can scroll through the tilt sequence using the slide bar at the low part of the window monitoring the centered patterns at the right hand side 21 60 interaction mode 1 view zool wing Image 1 1 r m Sa lh i 1 i ul ul i ll I l ne N IL hh il Il lll ll lll i i ll At the bottom of the shifted tilt series area there is a button selecting the interaction mode The default interaction mode is 7 zooming Diffraction patterns can be zoomed in using the right mouse button If some patterns appear shifted wrongly using the automatic procedure this can be corrected by manually setting the central beam position First it is recommended to zoom a diffraction pattern to ensure the correct positioning of a marker Then the interaction mode should be changed to 2 mark beam center In this mode the user can set the pattern center by left button mouse click on the appropriate position Once a position is selected a yellow cross appears at the point If by error several positions are marked the last given will be taken as active This active position is listed in the manual center points tab In order to take the select
9. difficult to measure the semi convergence angle It appeared that the convergence and therefore the effective camera length are strongly correlated with the settings of the diffraction lens used to additionally focus the pattern Therefore a fine camera length calibration procedure can be performed correlating the diffraction lens excitation and the effective pixel size Below a plot showing dix of a selected reflection plotted against the different values used for diffraction lens excitation is presented These values show a linear trend in a large region Therefore now knowing the nominal camera length and diffraction lens setting for an arbitrary pattern the correct pixel size in 1 nm can be found using the linear trend 0 216 0 214 0 212 0 210 4 S di 0 208 o 0 206 0 204 29 30 31 32 JJ 34 35 36 of 38 Diftraction lens It is handy to perform these calibrations for different camera lengths and fit in linear trend lines The plots should correlate the effective pixel size in l nm with the diffraction lens current The data is then can be arranged into a table as shown below camera length mm gradient y cut off 52 53 250 0 0367 2 683 300 0 0314 2 198 380 0 0279 1 581 560 0 0192 0 936 750 0 0157 0 654 1000 0 0087 0 604 ADT3D package has a possibility to perform fine calibration based on this table The data has to be measured for the microscope used and input into the fineCalibratio
10. fraction of clusters are sliced by the by the resolution limiting sphere and therefore they do not hamper the unit cell vectors determination Radius variation plot Since the two clustering parameters radius and min points are inter dependent one is varied while keeping the other constant Radius plot shows distribution of difference vectors while keeping min points fixed in this case 20 and varying the radius For each point while keeping the constant min peaks the radius is varied until the point falls into a neighbourhood of a certain cluster based on the assumed min points criterion It is seen from the plot that a lot of vectors can be assigned to a cluster with the neighbourhood radius of approximately 1 This region of the plot represents dense regions of the difference vector space Le points likely belonging to clusters The outer region of the plot covers the points which need a very large radius to fall into a cluster group Therefore these points are attributed as noise The proper choice of the radius for the selected min points is then given by the 34 BH Minimum points plot Lel eles u E 5 a E 3 E E 3radius dat 4 4 L 4 6000 8000 10000 12000 14000 difference vectors 35 inflection point around 1 Minimum points variation plot Minimum points plot shows distribution of difference vectors while keeping radius fixed in this case to 3 and varying minimum points
11. loaded a number of 2d functions become available Loading or creating a volume file MRCVOLUME type activates 3d functions in the menu bar Multiple MRC files can be loaded simultaneously However because the 3D volume files can be very large the developers advise loading only the necessary files for processing one data set at a time This will help avoid any memory capacity issues 11 12 6 2D viewer 2D viewer becomes active once a tilt series is open in the program 2D viewer allows visualization of the recorded diffraction patterns Within the 2D viewer additional navigation possibilities are available like going to the first lt lt previous lt next gt last gt gt patterns as well as resizing of the visualization window Original Size Fit to Window Diffraction data is visualised on an 8 bit scale additionally features such as histogram linearization HistEqual display the image with the highest available contrast The HistEqual mode can be switched off by clicking on the diffraction pattern area Ot Basic Image Viewer ice C fa ot Basic Image Viewer nce t a lt lt lt gt gt Original Size Fitto window HistEqual Invert 3D plot 3D plot boundingbox Rocking Curve lt lt lt gt gt gt OriginalSize Fittowindow HistEqual Invert 3D plot 3D plot boundingbox Rocking Curve H Showing Image 1 121 Showing Image 1 121 3D plot button opens an extra window with the curren
12. menu The binning is especially useful prior to the reconstruction of the reciprocal volume The binning procedure creates a new MRC file with the suffix _bi The binning is done in the additive way like it is done on a CCD chip by simple addition of 4 neighboring pixel values The newly created file can be saved by either using the Save MRC file icon or File gt Save MRC file menu At this point if the original file is not needed for the processing anymore it is recommended to close it in order to free memory space 13 14 7 2 Pattern centering During the tilt series collection the position of the central beam and therefore the complete pattern may shift The shift may occur for a variety of reasons one of the most common reasons is the diffraction pattern shift associated with the image shift which would occur when tracking the crystal during data acquisition It can also be caused by occasional use of diffraction shift knobs during the tilt acquisition The centering function is called either by pressing the Pattern centering icon or by clicking 2D functions gt Pattern Centering menu For each diffraction pattern of a tilt series the centre of the pattern is found and the pattern is shifted in order to bring it exactly in the middle of the frame There are several options of pattern centering procedure The default method of patterns centering is based on the analysis of the central beam The routine is launched
13. unit cell vectors 33 9 5 Clustering parameter tuning statistical approach 34 The clustering parameters can also be selected using a systematical approach The behaviour of the parameters can be visualized using Options gt Visualization of the menu bar Three different plots are allowed for visual inspection Visualization V Show parameter determiniation for Eps v Show parameter determiniation for minPts Z Show duster size distribution eme Below all statistical plots are presented and described Cluster size distribution bele lee gt gt Cluster size distribution hid 7 i ClusterSizeDistribution dat 0 size of cluster 0 10 20 30 40 50 60 70 cluster 20 minimum points nn T J0minimumPoints dat 6000 8000 10000 12000 14000 difference vectors Cluster size distribution plot On the horizontal axis all found clusters are presented So in this case 69 clusters were found The clusters are sorted according to their size in voxels The size is plot on the vertical axis This plot should show a homogeneous smooth monotone trend The smallest clusters appearing at the beginning of the plot are those clipped by the difference vector resolution limiting sphere Only a small
14. z 60 4 60 E E E 40 1 40 E 3 ial i 2 8 2 20 4 E 20 4 0 1 2 3 4 5 6 7 B 9 10 11 12 13 14 15 0 1 2 34 5 6 7 B 9 10 11 12 13 14 15 Displacement pixel Displacement pixel displacement plot 100 indicated refecions EEEE ins not indicated reflections mmm 2 B g F 60 4 E 40 4 8 Y 20 0 i j 0123 4 5 6 7 8 9 10 11 12 131415 Displacement pixel hki displacement plot 100 r indicated refiections mm wan not indicated reflechons ON s BO 5 zo 60 3 g 4 0 8 N A 20 0 0123 45 6 7 B 8 10 11 12 13 14 15 Displacement pixel 42 43 Together with the evaluation charts a window showing the reflection intensity appears In this window h k I Intensity and sigma estimated as a square root of the intensity value are represented intensity sigma BG o 2Dx o 2Dy r2Dx r2Dy angle boxsizex boxsizey px used ellipsex ellipsey ellipse 75490 274 754 x x x X x x x x x x X x 2939 171 441 x X X X x x X x x x x 8537 97 5688 x X X X x x X x x X X 20539 37466 47865 104263 143 314 193 561 x 218 781 322 898 4870 69 7854 40725 201 804 35478 188 356 6672 81 6823 46421 215 455 90034 300 057 19806 140 734 171477 414098 In order to save the hkl file the user should select the save standard HKL file option from the File menu of the main window BH ADT 3D Analysis 1 3 2 build GD Fie view E Open MRC file Save MRC file 2D functions
15. 121 16 TNT TRACTOR AT RTT ATA eT 1 Tal Tl VTE TA EAA AAT same AAA cr A l Il Hl MLL Ii ll HI IN Lill Will Hull MAAARI IN Ill Il Usually the approach of the shift vector propagation through the complete tilt series is a valid strategy for manually collected diffraction data The patterns typically do not move much as long as a pattern was not intentionally or erroneously shifted by the user during the data acquisition In this case an additional Friedel pair can be selected at the slice corresponding to the first pattern of the new shift vector and the centering procedure run as described with the use beam stopper mode checked If no beam stopper mode is used and manual center points as well as Friedel pairs are present in one pattern both being checked manual center points will override the Friedel pair centering Once the user is satisfied with the centering results accept button should be selected at the bottom right area of the window A new MRC data stack is created which is reflected in the Data Inspector field on the left The file name is extended by a _sh suffix If necessary this file can be saved otherwise it will be used for further processing and deleted after the program will be closed 15 16 7 3 Tilt axis determination rotation to the tilt axis To ensure accurate 3D reconstruction of the data the precise azimuthal position of the tilt axis within electron diffraction patterns is necessary Cautio
16. 44 aP 0 a 5 50813 b 12 7007 c 14 0843 alpha 101 872 beta 90 4664 gamma 102 145 44 aP 0 904623 a 5 50813 b 12 7362 c 14 0843 alpha 77 9551 beta 89 5336 gamma 77 1331 31 aP t Bms 15 3489 a 12 7007 b 5 50813 c 14 0843 alpha 90 4664 beta 101 872 gamma 102 145 35 mP 15 8952 a 24 8334 b 5 50813 c 16 903 alpha 80 7345 beta 123 974 gamma 90 3789 27 mC 29 2557 a 28 658 b 5 50813 c 12 7007 alpha 102 145 beta 104 043 gamma 79 3854 37 mC 37 4327 a 5 50813 b 14 0843 c 12 7007 alpha 101 872 beta 102 145 gamma 90 4664 34 mP 37 885 a 5 50813 b 24 8334 c 14 0843 alpha 77 7462 beta 90 4664 gamma 90 3789 38 oC 45 743 a 5 50813 b 24 8334 c 28 658 alpha 78 0331 beta 100 615 gamma 89 6211 26 oF 51 1454 a 16 9136 b 20 8153 c 5 50813 alpha 82 943 beta 99 4829 gamma 83 9576 25 mC 51 5187 a 5 50813 b 12 7007 c 14 0843 alpha 101 872 beta 90 4664 gamma 102 145 33 mP Below the centered unit cell vactors are ploted within the difference vectors of the centered lattice the thrue vecors do not point to a cluster 39 40 Even if the lattice is centered it is recommended to first index refelctions using the primitive lattice and then recalculate the indices of the refelctions within hkl file Once the correct centered cell 1s found it can be sent to the 3D viewer it can be saved Subcequent indexation of the refelctions will be done using the currently selected unit cell basis 40 41 12 Extract
17. For each point within a fixed radius number of points is calculated The initial region of the plot corresponds to low density regions where within the defined area only a few points are met After the inflation point the plateau is observed representing area of true clusters a lot of difference vectors having a similar high number of min points sitting inside the defined radius Therefore for the clustering procedure the number close to the inflection point should be taken It should be noted again that the two parameters radius and min points are dependent and changing them will change the behaviour of the plots 35 36 10 Difference vector space viewer Difference Vector Space Viewer visualizes the part of the difference vector space used for the unit cell vectors determination Difference vectors are shown as grey points if they are not assigned to a cluster and threated as noise and coloured in random sequence if they are assigned to a cluster Three automatically defined unit cell vectors are shown as green b and blue c lines crossing at the coordinate origin The File menu of the Difference Vector Space Viewer offers possibilities to save the found cell file and send the cell to 3D viewer The text field shows the unit cell parameters for the current cell vectors The vector lengths are shown in pixels for the reciprocal space vector set and in Angstroms in the direct space if the units are set to Angstroms
18. at the intensities can range from O to 65 535 MRC files contain 16 bit signed data If within the original tif file the intensities are below 32 767 the data can be directly transformed into the positive part of the MRC file If the counts are above the 32 767 level two possibilities are offered by the program divide by 2 or truncate r Packing options Typically high counts belong to the central transmitted beam or originate from hot spots and these intensities are not important In this case the counts higher than 32 767 can be neglected set to 32 767 using the truncate option If the user decides to keep the intensity ratio also for the strong counts all number can be divided by 2 in order to fit into the 0 32 767 range Negative counts are allowed by should not comprise a major fraction of a pattern this may cause problems in intensity integration procedure After all diffraction patterns are packed in the the 3D data stack the essential information about the experimental data is written into the MRC header The user is prompted by the program to input the relevant values e Starting tilt position please use negative to positive tilting direction If you collected the data in the other way still use the positive tilt step direction at the end it will only flip the data e Tilt step should be positive see above Only equidistant tilt positions are supported at the moment e Pixel size in 1 nm e
19. ata quality and quantity to the field of electron crystallography and greatly improves the chances of success structure solution from electron crystallographic data ADT3D is intended to make the process of data extraction from raw diffraction images to hkl intensity file for structure analysis as quick and easy as possible For an experienced user using a modern PC with ample memory capacity we recommend 4 GB and a reasonable data set quality this process can be as fast as 15 minutes Our wish is that ADT3D will help make the ADT technique accessible to as many researchers as possible even those who would not have dared to attempt electron crystallography in the past Our guiding philosophy has been to make the user interface and experience as simple as possible while remaining flexible and adaptable for the user to deal with challenging data sets We are grateful for any feedback which can improve the program The ADT3D package contains the following Tiff 2mre converter ADT3D Example data sets Barite Natrolite GnuPlot 4 2 Thank you for purchasing our software The ADT3D development team Centre for High Resolution Transmission Electron Microscopy Institute for Physical Chemistry Johannes Gutenberg University Welderweg 11 55099 Mainz Germany Institute for Informatics Staudingerweg 9 55099 Mainz Germany The program is distributed via NanoMEGAS SPRL Blvd Edmond Machtens 79 B 1080 Brussels BELGIUM 2 In
20. ble dynamic range of the CCD should be used the reflections should not be saturated this would spoil their intensity but they also should not be too weak The central beam can be saturated Gf the CCD can tolerate it e For an attempt of a complete structure analysis largest possible tilting range should be covered 60 or even larger for unit the cell parameters determination a much shorter tilt sequence can be sufficient 30 or less depending on the data quality e The success of a structure analysis procedure is especially enhanced when precession electron diffraction is used especially for inorganic materials e The use of a beam stopper to blank the transmitted beam is tolerated by the program as long as there are some diffraction patterns with clear Friedel pairs and the shift vector of different patterns in the tilt series is similar see the centering routine 4 TIF2MRC data converter At the moment the program can only read electron diffraction tilt series as MRC files the file format description can be found in the specified appendix MRC is a standard file format for 3D data used in tomography and there are plenty converting possibilities existing IMOD 2dx ImageJ etc We nevertheless recommend to use the converter delivered together with ADT3D for the following reasons electron diffraction patterns typically and hopefully have a very high dynamical range which should be treated in a certain manner when convertin
21. d in a lattice but each node of the lattice is smeared out Data clustering should group vectors into sets corresponding ideally to each node In ADT 3D package the clustering procedure is done using the following algorithm Two parameters are used to cluster the data radius and min points Below the definition of the parameters is presented schematically Radius defines the size of the neighborhood min points tells how many points should present within the neighborhood for the point to be assigned to a cluster O Each data set may have different distribution of the points given by size of the unit cell distance between the clusters true reflection shape crystal bending etc These parameters should be adjusted for each data set Using too large radius may lead to fusion of separate clusters too short radius may lead to cluster splitting If min points is set to a too high value it may happen that no clusters will be found at all as the cluster definition is not satisfied If the value is set too low only a few large clusters will be found likely consisting of several lattice nodes It is strongly recommended to change radius and min points parameters separately as they strongly influence each other Refine cluster option was introduced in order to make the cluster more clear once they were found correctly Data points assigned to a cluster but placed on the outer shell of the cluster 30 31 are omitted using the given thre
22. e view on your data On the left hand side you see the default settings where the complete value range 0 5699 is transferred to 0 256 for viewing Since in this example almost all values over 2000 are null you can reduce the right border of the range using the right mouse button This will map the new range in a better way to 0 256 and you will get more contrast in the 3d space AOT zo Histogram AOT zb Histogram real range 5699 real range 5699 E EN r Y U l o i A Left Border Right Border 75 a h A Left Border Right Border 255 2 Logarithmic View Logarithmic view Experimental Data gt Volume Rendering gt Adjust gt Transfer allows you to change the hue saturation value or alpha value for each map value 0 256 First try to change the alpha value If you see all the necessary information enhance the visualization change or add new hue values _3D 3 ss0ff_bk_sh_ro_provad5 2999_rotref 0 5_bi_dawnsampled_dow DX so Transfer Function PR 20 21 Found reflections The Found reflections drop down menu offers the visualization options choice for the reflections The two first parameters min distance cut off and max distance cut off are related to the find reflections parameters ADT 3D Analysis 13 2 bu ild 2704 3D 4 BaSO4 PED_sh_rot45 bi re bi rel elle Qt File View Data reductio 2D functions 3D functions Workspace Options Help E X PA RH
23. e distance between the two selected clusters and the angle between the corresponding vectors When the third vector is selected parameters of the unit cell build on these three vectors are shown in the text box Any three selected vectors can be defined as a new basis set by pressing the tools lt set cell button Pressing the tools lt find cell button recalls the automatic cell determination routine for the available cluster set and delivers the initial cell vector basis A selection of tools is provided in the tools drop down menu Tools menu of the Difference Vector Space Viewer can be best used in the editing mode The origin of the basis vector set can be shifted to any cluster by selecting the cluster and choosing Tools gt Centring gt Set to Centre option The back sift of the origin is achieved by Tools gt Centring gt Centre to Origin command A list of options allows adding reciprocal lattice nodes to the existing cluster set All these functions are available from Tools gt Add Clusters menu 1 2 creates a node in the middle between two selected existing nodes 1 3 creates two nodes between two selected existing nodes spaced equidistantly double creates a node with a double distance from the lattice origin along the direction given by a selected node invert creates a node with the coordinates opposite by sing to a selected node add family adds a nodes set to the cluster set with the or
24. e menu 2D functions gt Interpolate 3D command A volume MRC file is then produced with the suffix _re originating from reconstructed 18 19 8 The 3D viewer For the reconstructed volume file 3D functions become available 3D viewer find peaks find cell and integrate reflections The 3D viewer opens four windows with the volume data projected along different directions In the menu bar of the 3D viewer a user can select Experimental Data Found reflections Cell and View functions BH ADT 3D Analysis 1 3 2 build 2704 ECE Qt File View Data reductio 2D functions 3D functions Workspace Options Help LEX J ee m oO R EN CPB E PH ASS Data Inspector E X Experimental data Found reflections Cell View Properties Area x 4 detailed dataset info gx Qt 4 BaSO4 PED_sh_rot45_bi_re x info MRCHeader Cell TYPE MRCVOLUME BaSO4 TYPE MRCVOLUME File name PED_sh_rot45_bi_re mrc a BaSO4 PED_sh_rot45_bi_re mre Volume 512 x 512 x 444 2 A_Tatiana BaSO4_PED BaSO4 PED Memory 222 MB pixel voxel size 1 170 2 A 4 edit change cell informations Gx Cell Nr 1 Cell Nr 2 Space information pixel angstrom Cell parameter Dimensions 170 202 170 202 170 202 zD ND speer l naia m Log ax RsmMRCFileHandler writeMRCFile Completed in 57 sec SUCCI i Completed in 56 796 seconds open Axis3DViewer Each of the windows show a different project
25. e prior usage determined that certain values were fixed too low and the addition of more types dictated the necessity for numbers below while O is generally avoided The 0 is typically an undefined number Numbers for microscope aberration between parentheses give the level of the aberration Name a tilt b tilt X stage y stage z stage x shift y shift Byte position 0 4 8 12 16 20 24 Comment Alpha tilt in degrees Beta tilt in degrees Stage position in um Image shift position in um 48 defocus exposure time mean intensity tilt axis angle pixel size magnification Microscope type Gun type Lens type D number of microscope High tension kV Focus spread diffraction lens value MTF Starting Df Focus step DAC setting Spherical aberration Semi convergence Info limit Number of images Image number in series Coma 3 X Coma 3 Y Astigmatism 2 X Astigmatism 2 Y Astigmatism 3 X 28 32 36 40 44 48 52 56 60 64 68 12 76 80 84 88 92 96 100 104 108 112 116 120 124 128 49 Defocus in nanometers note in MRC standard it is micrometers Image pixel size in nanometers Tecnai 10 2 Tecnai 12 1 Tecnai 20 1 Tecnai 30 2 for Titan add 10 to HT equivalent W 2 Lab6 1 FEG 1 HC 3 BioTWIN 2 TWIN 1 STWIN 1 UTWIN 2 XTWIN 3 CryoT WIN 4 Focus spread for Truelmage acquisition Diffraction le
26. eclection centers geometrical mean coordinates and weighted geometrical mean values 2 22 Show reflections full Max value 22 Cell 23 The Cell drop down menu sets the visualization options for the unit cell vectors once they are found Two cells can be shown simultaneously using the second cell display option Cell coordinate system option visualizes cell basic vectors color coded as a red b green and c blue Multicell option allows visualizing a reciprocal lattice built using the defined vector basis The number or repetition vectors in each direction can be specified using the configure multicell option ADT 3D Analysis 13 2 build 2704 3 Qt File View Data reductic 2D functions 3D functions Workspace Options Help gt m oS GA LH 8 D B E pH MAA Data Inspector B X Experimental data Found reflections Cell View Properties Area EX vV Cell Displa 4 detailed dataset info a x U 4 BaSO4 PED sh rot45 bire ollie ra SECOND Cell Display info MRCHeader Cell TYPE MRCVOLUME Cell Coord System ame BaSO4 MultiCell TYPE MRCVOLUME PED_sh_rot45_bi_re mre config MultiCell BaSO4 Fil 1 PED_sh_rot45_bi_re mrc Volume 512 x 512 x 444 2 A_Tatiana BaSO4_PED BaSO4 Path PED Memory 222 MB pixel voxel size 1 170 2 A 4 edit change cell informations Gx Cell Nr 1 Cell Nr 2 Space information pixel angstrom Dimensions 170 202 170 202 170 202 ax 23
27. ections indexed sorted according to the volume of the reflections Here the indexing procedure performed a medium quality job less than 70 of total number of found reflections was indexed color coded yellow from 60 to 80 bar Among them very small reflections with the volumes fewer than 40 voxels were indexed particularly poor only 60 out of them were assigned an index On the contrary the largest reflections with the volume of more than 80 voxels more than 80 of reflections were indexed color coded green Indexed he a K n K 1 ar G ee F LK gv Percent 15 610 11 20 21 40 41 60 gt 80 Al Volume of reflections Al 42 Different displacement plots are shown for the indexed reflections The prots show the indicated and not indicated reflections plot versus the displacement In the plots below most indexed reflections have short displacements from the exact position of the lattice node The threshhold for this displacement was defined in the indexing parametr menu Generally at this point the user should take care that most of reflections are indexed especially those placed close to a lattice node Here the user should observe a fading plot showing that there are only few reflections with a high displacement h displacement plot k displacement plot indicated refiechons mmm not indicated reflecions mm indicated refiecions EEEE not indicated reflections mm 80 ZD a T 5
28. ed positions into account in the general tab manual input selection area use manual central points box has to be checked Then the run centering button will call the centering routine which would center all patterns based on the transmitted beam shape expect for the patterns with the yellow marked central points 14 15 If during diffraction data collection a beam stopper was used the user cannot set the position of the central beam precisely In this case for the centering symmetry equivalent reflections Friedel pairs should be used The interaction mode should be changed to 3 mark Friedel pairs Then a pair of symmetry equivalent reflection should be selected by left button mouse click a green cross appears at the marked position Simultaneously a list of marked reflections appears in the Friedel pairs tab In the general tab use manual Fridel points box has to be checked With this set of parameters all patterns will be centered using the central beam except for the pattern where the equivalent reflections we selected In order to propagate the shift vector from one frame to the complete tilt series use beam stopper mode should be checked Now having the beam stopper mode active and only one Friedel pair selected the run centering will shift all patterns using the same shift vector calculated from the Friedel pair a Show interaction mode 3 mark Friedel pairs TAT HINA A jing Image 45
29. ee ngle G gt Volum Co Ba t 4 x Y zZ ist Vect 276 2nd Vect 4 S 3rd Ve AD cof Scale spe Scale 0 5 ata e w Search of reflections started Number of reflections found 985 finished in 2 seconds The next processing step includes calculation of difference vectors for the found reflections The number of difference vectors can be estimated as n of the number of the reflections found Generally 1000 1500 reflections producing around a million difference vectors should be enough to deliver unit cell vectors Therefore if the number of found reflections is too large the user should go back to Find reflection parameters panel and either increase the min volume value or the threshold value Both actions are working in the same direction reducing the number of reflections Reflections are found as sets of adjacent voxels with a total intensity above a certain threshold In principle there are different ways to calculate the coordinates of these 26 21 reflections just taking the voxel with the highest intensity out of the set taking geometrically mean value of the set or weighted through the intensities mean value 3D viewer offers displaying reflection centres found according to these different approaches by selecting the appropriate option in Found reflections gt reflection centres menu Normally these coordinates do not differ significantly Sometimes several strong reflections are placed one next to eac
30. er called ADT3D somewhere e g C Program Files ADT3D or D ADT3D on your system where you have a least 200 MB of free disk space 2 Change the access rights of this folder that every user can read write and delete into this folder 3 Copy or extract the ADT3D files into this new folder 4 Make a shortcut to the ADT3dAnalysis exe available for all users on their desktops the easiest way to do it is to put the icon on the desktop of the public user Start ADT3D keep tabs on the starting process and test the 3d viewing and your graphic card via File gt 3d test 3 A few words about the ADT experiment In ADT experiment electron diffraction patterns are collected through a tilt around the primary goniometer axis Usually the patterns are collected using equidistant tilt steps although principally arbitrary tilt steps can be used at present the TIF2MRC converter can only deal with the equidistant tilt steps but can be adopted on demand The ADT experiment can be done either using nano diffraction or selected area diffraction The operator must ensure that the diffraction patterns are collected from the same crystal Other crystals eventually appearing in the field of view may disturb the analysis but are not crucial if there are only a few Several points must be taken into account while making the experiment e The operator should take care that the patterns are roughly centered on CCD e The possible largest fraction of the availa
31. eshold can be set within the Find reflection parameters panel When the options of the reflection search procedure are defined the procedure can be started by clicking on the Find Peaks icon of the main window menu Experimental data Found reflections Cell View show min dist of cut off show max dist of cut off show refelections full reflections centers a A In the bottom log panel of the main window the number of reflections found is displayed For the case presented 985 reflections were found When the show reflections full option is switched on reflections are shown in the 3D viewer with blue markers BH ADT 3D Analysis 13 2 build 2704 DDA Ba Doh rot45 bi re Safar x Qt File View Datareduction 2Dfunctions 3Dfunctions Workspace Options Help Sx 2 gt EER LH 2 E DRB amp 8 PH MAA Data Inspector X Experimental data Found reflections Cell View Properties Area ax 4 detailed dataset info x CI 4 Ba504 PED_sh_rot45_bi re olle TYPE MRCVOLUME BaS04 PED_sh_rot45_bi_re mrc info MRC Header Cell TYPE MRCVOLUME File name BaSO4 PED_sh_rot45_bi_re mrc Volume 512 x 512 x 444 Z A_Tatiana BaSO4_PED BaSO4 PED Memory 222 MB pixel voxel size 1 170 2 A 4 5a mea j rar Q p y 4 edit change cell informations x 2D 5 Cell Nr 1 Cell Nr 2 pi ste A KO d d md Space informati nse E pixel angstrom D qa t 8 Cell parameter g ae eh oer
32. g the Find Cell icon of the main window starts the automatic unit cell parameters determination based on the found set of peaks After the procedure is finished a Difference vector space viewer window opens showing a half of the difference space volume the other half is symmetry equivalent and found coloured clusters within the volume The centres of detected clusters are marked by small red spheres These spheres represent nodes of the reciprocal lattice Three shortest non coplanar vectors towards cluster centres are found automatically and are marked as a red b green and c blue v limit resolution related to shortest cell length Distance 51 Neighborhood definition clustering Find cell options can be set in Options gt Cell calculation parameters menu of the main window The length can be represented either in pixels or in Angstroms Difference vector definition parameters allow selecting between the found reflection sets the user can use either the reflection centres defined by the maximal value within the reflection geometrical centre of mass or weighted centre of mass The shortest difference vectors correspond to the shortest distances in the reciprocal lattice Very short difference vectors are usually omitted because they are normally artifacts of calculation The cut off can be set by activating the cut off centre option together with the input of the threshold Difference vectors are calculated between all reflect
33. g to signed 16 bit MRC format and essential experimental information should be written in the heater which is typically not used in image based tomography As default TIFF2MRC converter is delivered EMI2MRC TIA and DM32MRC Digital Micrograph possibilities also exist and can be compiled into a stand alone program on demand Installation of the Converter e Copy the ADT_tiff2mrc_pkg exe to a new folder and make sure to have installation administrator rights e Double click on the ADT_tiff2mrc_pkg exe follow the instructions e Double click on ADT_tiff2mrc exe to run the converter program Prior to the converting procedure the input TIFF files should be sorted according to their tilt angle position and have names numbered sequentially using the scheme Generic name counter tif For example the files can be named Baritel tif Barite2 tif Baritel0 tif Baritel l tif It is assumed that the files are sorted from negative tilt angle values going into the positive direction and the software makes the assumption that the tilt steps are always the same size i e the tilt positions are equidistant Once a diffraction pattern for a certain tilt position is missing it can be substituted by a blank TIFF file having the same size in pixels as the other diffraction patterns and the corresponding name When the converter is started the user is asked to pick up the first TIFF file of a stack The file name is analyzed the las
34. h other so the simple intensity threshholding considers them as a single reflections This may be problematic for the subsequent unit cell parameters determination procedure as these reflections will deliver false coordinates not belonging to the lattice Alternatively very small strong spots can be a refult of a hot spot thus also diminishing the data quality To exclude these reflections an additional option is available set maximal and minimal volume for a reflection 27 28 9 2 Manual assessment of the unit cell vectors The three reciprocal vectors are coded as red green and blue colours These correspond to the reciprocal unit cell vectors a b and c The information about the unit cell directions appears in the properties area right side under the header information As default the vectors are set to x y and z directions In the free rotation window upper right the user can select a specific orientation and set this orientation to the a reciprocal vector by pressing shift a button combination on the keyboard Consequently the volume in the lower left window red vector projection will orient along the selected direction Consequently in the base vectors coordinates list the first vector coordinates are changed Applying the same strategy to the b and c directions rotate within the free rotation window and press shift b shift c the user can define three vectors of the unit cell The vectors coordinates in pixels are pre
35. idual images e Image data Note There are two versions of the file format The shorter version is not completely adhering to SI units meters radians the longer version has as added value whether SI Units are adhered to The primary header Important values for TrueImage are high lighted Data type Name Byte Comment position integer 4 byte nx 0 number of colums integer ny 4 number of rows integer nz 3 number of sections 1 e images integer mode 12 data type pixel value stored as short integer integer nxstart 16 lower bound of colums integer nystart 20 lower bound of rows integer nzstart 24 lower bound of sections integer mx 28 cell size in Angstroms pixel spacing xlen mx integer my 32 integer mz 36 4 byte floating point xlen 40 4 byte floating point ylen 44 4 byte floating point zlen 48 46 4 byte floating point 4 byte floating point 4 byte floating point integer integer integer 4 byte floating point 4 byte floating point 4 byte floating point 2 byte integer 2 byte integer 4 byte integer 2 byte integer char byte 2 byte integer 2 byte integer 2 byte integer 2 byte integer 4 byte floating point 4 byte floating point 4 byte floating point 4 byte floating point alpha beta gamma mapc mapr maps amin amax amean ispg nsymbt next dvid extra 30 numintegers numfloats sub Zfac min max2 min3 max3 52 56 60 64
36. igin on a selected cluster undo all removes all newly created nodes leaving only found clusters oy 38 38 39 11 Transformation of the unit cell The three basis vectors of the unit cell are found automatically based on the cluster set as three linear independent vectors with the shortest distance from the origin Niggli cell reduction procedure is performed on these vectors In the example presented below the primitive unit cell has a metric of 14 08A 12 70A 5 50A 102 1 90 5 101 9 Based on this metric a higher symmetry than triclinic is possible Possible cell transformations can be found in the Tools lt analysis lt conventional settings drop down menu BH Difference Vector Space Viewer a x Possible cells are listed sorted according to the figure of merit based only on the unit cell metric no reflection symmetry is infolved at this point The first set of vectors has a figure of merit O this is the cell as it originally found by the programm The next line represents the triclinic cell in the standard settings with the lengths of the vectors sorted and all angles under 90 The following line shows a centered monoclinic unit cell with the lattice parameters of 24 83A 5 51A 14 08A 90 5 102 3 89 6 Thus the final cell has a monoclinic metric and C centered lattice BH List of conventional cells o a 5 50813 b 12 7007 c 14 0843 alpha 101 872 beta 90 4664 gamma 102 145
37. ion of HKL Intensity data set Once the unit cell basis vectors are defined the set of found reflections can be indexed and intensities assigned to reflections can be extracted A lattice is built based on the basis cell vectors When a lattice node comes close to a found reflection it is given the nodes indices The intensity is calculated as a sum of all voxel intensities assigned to the reflection Note at the moment only found reflections are integrated So after you found unit cell vectors for a selected set of reflections you may consider researching for reflections in order to include also weak spots for subsequent integration The most critical parameters for the reflection indication and therefore proper integration can be found in the options lt indexing parameters menu In this menu the user can set the acceptable indexing error for each crystallographic direction in terms of fraction of the unit cell vector length The default values are 15 in all three vector directions v use same center type used for cell determination Center type mean value Positioning error a b el 15 cii 15 s The indexing and intensity integration procedure can be called by pressing the Integrate reflections HKL icon of the main window menu Two windows pop up after the integration procedure The first window represents the evaluation charts of the indexed reflections The top left chart Indexed shows percentage of the refl
38. ion of the diffraction volume Orientation of the volume within upper left lower left and lower right windows are predefined directions The volume shown in the upper right window can be rotated freely using the left mouse button Within the preset orientation windows the image can be zoomed by clicking the left mouse button and while keeping it pressed moving the mouse towards the centre of the image or in the opposite direction Alternatively the images can be zoomed by rotating the mouse wheel Within the non orientation preset window upper right the image can only be zoomed by the mouse wheel The user can toggle between the four windows view and one large window view by placing the mouse cursor over the desired window and pressing the space bar of the keyboard Experimental Data The Experimental Data drop down menu offers the visualization options choice for the 3D data ADT3D supports volume rendering which can be configured via the Experimental Data gt Volume Rendering menu For each dataset a transfer function grey values or colored can be saved loaded or changed The default transfer function is grey scaled and should fit mostly all experimental dataset Via Experimental Data gt Volume Rendering gt Edit gt Load Save Transfer Function TF you can load some other transfer functions or save your own transfer functions 19 20 Experimental Data gt Volume Rendering gt Adjust gt Windowing allows you to change th
39. ions Some of them may be very long Working with all difference vectors is a very computationally intensive an unnecessary task Therefore only difference vectors within a certain shell are considered further for unit cell vectors calculation The size of this shell is defined by the limit resolution distance either 1n pixels of in Angstroms This parameter is very important for the unit cell vector determination Below a situation is demonstrated when the distance is set to a too low value left side The increase of the distance middle and right images covers larger volume and therefore gives more flexibility for the unit cell vectors determination 29 30 Two parameters are used for the clustering routine analysing the different vector space combined in the neighbourhood definition group radius and min points As the reflections are not distributed homogeneously but sit mostly on the nodes of a 3D lattice the difference vector space must have the same periodicity but strongly enhanced especially for short vectors In a way the difference vector space can be considered as an autocorrelation of the reciprocal lattice The three shortest non coplanar vectors found in the difference vector space should correspond to the primitive unit vector set of the original lattice Due to geometrical uncertainty of the reflection position slicing error during the data acquisition physical size of reflection etc difference vectors are organize
40. lt axis not far away here 10 from the correct axis and in c a clear lattice for the correct tilt axis You find all stereographic projection images in JPEG format in a folder called FILE NAME tilt angle search next to your MRC input file The result of your manual or automated search is shown in the plot below a i p tilt axis search results gnuplot image ssoff sh mre R44 46 Step 1 searchTiltAxis plots Actually we take the highest value of the variance profile If you have more than one clear peak in the plot or the maximum profile differs from the variance profile please start manual searches in all regions and compare the stereographic projections to each other After a tilt series 1s rotated to make the tilt axis vertical a new MRC file with the suffix _ro with the tilt axis position is created The images were taken from a strong BaSO4 dataset If you have few weak reflections the stereographic projection might be not so detailed but still accurate enough for the search process 17 18 7 4 Reconstruction of the volume data Centering and in plane rotation to the tilt axis steps prepare the data for the diffraction volume reconstruction The data volumes are often very large and can cause memory issues it is often appropriate to bin the data before creating a diffraction volume The reconstruction of the volume is done by pressing the nterpolate 3D icon or selecting in th
41. n it is assumed that the tilt axis position within a tilt series is the same for all diffraction patterns The procedure of the tilt axis determination is called by clicking the icon Refine Tilt Axis or from the menu 2D functions gt Refine Tilt Axis On the click a dialog window Search or rotate to tilt axis appears The position of the tilt axis is measured from the horizontal is a standard Cartesian coordinate system X to the horizontal axis and Y to the vertical i e 356 means 4 below the horizontal The tilt axis position can be stored and recalled from MRC header A user can freely select a reasonable value for the tilt axis and put it into the manual input box When the exact position is not known or needed to be refined search for the tilt axis routine has to be started A user is more flexible using the manual search routine where the whole search can be confined to reasonable ranges and searching steps Automatic search routine uses specific preset parameters and may take longer BY Search OR rotate to tilt axis You have the choice rotate to known tilt axis from MRC Header You can rotate to the angle given in the MRC Header MRC Header angle 272 3 Manual input You can rotate to a free known angle Use this to apply your search results your choice 272 3 Then the dataset is rotated by pressing START and a new rotated dataset appears search for unknown tilt axis manual SEARCH for tilt axis You can search ma
42. n be input either in pixels or given the correct pixel size in reciprocal Angstroms is available from the MRC header in Angstroms The area around the transmitted central beam in electron diffraction patterns is usually not reliable for reflection determination and therefore it is advised to exclude this area for the reflection search procedure cut off zero beam Activated show min dist of cut off option of the 3D viewer gt Found reflection menu marks the omitted area as a semi transparent red sphere around the central beam eo 2 i a SE E oe ee ee a ae ed seto 0e ee ese e EE Ee er 4 es am am kaa oF tad r e d id e 2 v v amp 3 se teaeteerg ie Per er ce eee oer eee Pere ass ese CK uaa Wee ee Ate eeen sarren sr Ass eee ss uma tree er asses wk ka s s OOO HHHMEHO HOT OO 6s e oa ess sa 38e oe ee et se srte ks tagen se eeens eve eee et es tegee e ee ee Lo ee g Sedes o 25 26 Certain tasks may require excluding high resolution in terms of diffraction areas during the reflection search procedure This is achieved by activating the limit resolution box within the main window menu Options gt Find reflection parameters panel The used resolution range is then green colored Reflections are defined as objects having more than min volume adjacent voxels and having total intensity higher than a certain threshold The values for the min volume and the thr
43. n txt file which should be in the same folder where the main executable ADT3D file is When the fine calibration file is not available the pixel size in 1 nm read from the MRC header will be used The information about the diffraction lens current can be stored in and recalled from MRC header 53
44. ns value in percent for diffraction tomography nm in non SI M in SI nm in non SI M in SI mm in non SI m in SI mrad in non Si rad in SI nm for non Sl m for SI nm in non SI min SI nm in non Si min SI nm in non SI m in SI nm in non Si min SI nm in non SI m in SI 49 Astigmatism 3 Y Camera type number Camera position Spherical Aberration 4 Star Aberration 4 X Star Aberration 4 Y Astigmatism 4 X Astigmatism 4 Y Coma 5 X Coma 5 Y Three Lobe 5 X Three Lobe 5 Y Astigmatism 5 X Astigmatism 5 Y Spherical Aberration 6 Star Aberration 6 X Star Aberration 6 Y Rosette Aberration 6 X Rosette Aberration 6 Y Astigmatism 6 X Astigmatism 6 Y SI Units 132 136 140 144 148 152 156 160 164 168 172 176 180 184 188 192 196 200 204 208 212 216 nm in non Si min SI WA 1 MSC 2 GIF 3 m S i to eS wes i iS we wee 0 if false 1 if true 50 50 51 13 3 CELL file format At the moment two different cell file format are supported binary and ascii The binary cell is a 3x3 matrix with pixel coordinates floating point 32 bits precision in reciprocal space in the a a a X y Z format b b b The cell coordinates in the ascii file format are much more intuitive and can be read by user directly The format is ADTCELLVERSION 2 reciprocal space cell vectors pixels ar a
45. nually for the correct rotation Here after computing a while the result appears and you can insert the correct value above or start a new search You can specify the search range as well as the step size to find the best tilt axis Search Range 90 30 Search Stepsize 5 automatic SEARCH for tilt axis You can search automatically for the correct rotation Here after computing a while the result appears and you can insert the correct value above or start a new search You can specify the angular accuracy in degrees The automatic search routine uses one or more than one iteration to find the best tilt axis After each iteration the search intervall is reduced around the best value and the step size is adaped automatically angular accuracy in degree 0 1 The search for the tilt axis 1s based on the assumption that for the correct tilt axis value correct diffraction geometry for 3D reconstruction the reciprocal volume should contain a coherent 3D lattice The coherent lattice should produce a clear well defined pattern in stereographic projection If the position of the tilt axis 1s wrong the 3D reciprocal lattice is twisted and the stereographic projection is blurred Therefore the sharpness and the brightness of the stereographic projection can be used as criteria for the tilt axis search 16 17 The three stereographic projection above show in a a blurred image for a complete wrong tilt axis in b a ti
46. oe v el CBE i pH Ae Data Inspector E X Experimental data Found reflections Cell View Properties Area x 7 show min dist of cut off 4 detailed dataset info G x Qt 4 BaSO4 PED_sh_rot45_bi_re fo e show max dist of cut off info MRC Header Cell TYPE MRCVOLUME show refelections full dienam BaSO4 reflections centers max value re a TYPE MRCVOLUME j PED_sh_rot45_bi_re mrc SN z on BaSO4 r PED_sh_rot45_bi_re mrc Volume 512 x 512 x 444 2 A_Tatiana BaSO4_PED BaSO4 Path PED Memory 222 MB pixel voxel size 1 170 2 A 4 edit change cell informations Gx AN 2 Cell 1 celnn 2 AD cot Space information In pe pixel angstrom D gia Cell parameter Dimensions 170 202 170 202 170 202 Angles Volume Bast te x Y Z ist Vect 2nd Vect D 3rd Vect A L E or Scale ise ama Log g x RsmMRCFileHandler writeMRCFile Completed in 57 sec successful Completed in 56 796 seconds open Axis3DViewer EH Find Reflections O Dti ons Units pixel voxel angstrom Cut off v cut off zero beam Distance 26 limit resolution Distance 425 Find reflections Min volume 5 Max volume 100000 Threshold 66 manen nl neten Found elfections can be shown in different ways show reflections full option colors all areas assigned to as reflections The centers of the flectrion can be visualized using different positions center positions corresponding tohe maximal intensity values geometrical r
47. patterns Volume data An MRC file containing the 3D diffraction volume in Cartesian coordinates All diffraction pattern processing is done within stack data files while 3D operations like lattice vectors determination are performed within volume data The description of MRC file format used here can be found in the appendix Tilt series MRC files must have an extended MRC header containing the information about the tilt sequence pixel size diffraction camera length etc This information is essential for subsequent volume reconstruction and processing The following is a brief outline of steps used to go from raw images to extracted intensities e Load Data Assumes a MRC stack of your data is available otherwise it can be created from a stack of TIF images using TIF2MRC converter provided together with the package e Inspect in 2D viewer check for blank images or data which looks to be bad or corrupted Shift diffraction patterns to a common origin Find tilting axis Rotate data to the Tilting axis Bin data to save memory space 3D volume is only a visualization tool Create 3D volume Inspect data in 3D volume Find reflections default parameters estimated for your data should give roughly 1000 reflections Find unit cell optimize clustering parameters if necessary e Extract intensities In order to open an MRC tilt series click on File gt Open MRC file A dialog window allowing you to browse through your data will ap
48. pear Select the target file and press the Open button Alternatively the Open file icon can be used A progress window will appear as the data stack is being loaded 10 BH ADT 3D Analysis 1 3 2 build 2702 File View Datareduction 2D functions 3 Dfunctions Workspace Options Help OERS E Hsp AT e Data Inspector ex CJ 0 Ba504 PED cole TYPE TILT_SERIES File name BaSO4 PED mre TILT_SERIES BaSO4 PED mre 2 A_Tatiana BaSO4_PED BaSO4 Path PED 242 MB 1 416 8 A pixel angstrom Cell parameter Base vectors 3D ADS ocio ERAN nsP Scale rms 0 nlabl 1 labl 800 Once the data is loaded a Data Inspector window will appear on the left side showing the type of the data in this case TILT_SERIES file name image size 2D size in pixels of a single diffraction pattern image count the number of diffraction patterns in a stack and the tilt range read from the extended header On the right side a corresponding Properties window will appear with more detailed information the MRC header are summarized in the Properties area The info panel presents the information on the data type file name data path memory used by the loaded file and pixel size in reciprocal units within diffraction patterns The MRC header panel lists data available from the basic MRC header The parameters within the MRC header cannot be edited from the MRC header panel Once a file of TILT_SERIES type has been
49. sented in the cell information panel Within the reciprocal space panel the user can switch between the pixel and Angstrom representation of the vector lengths Cell parameter lengths in the reciprocal space and therefore consequently in pixels or in reciprocal Angstroms and the angles are presented in the panel These values of the vector lengths can be changed freely the coordinates of the cell vectors are adjusted accordingly The length of the three vectors can be scaled by a selected factor by pressing the Scale button The overlap of the created unit cell vector set and the diffraction volume can be visualized by pressing the Show button at the bottom of the cell information panel The unit cell can by displayed within the diffraction volume data as a single cell having the View gt Cell display option of the 3D viewer active or as a lattice by activating the View gt MultiCell option As soon as an appropriate unit cell vectors set is created it can be saved by clicking the Save CELL file icon in the main menu Cell vectors are saved in a file with the cell extension see Appendix for file format description 28 29 9 3 Automatic unit cell determination The approach described above demonstrated a manual cell parameter determination procedure based on the visual selection of the appropriate reciprocal space directions In this section the automated cell parameter determination routine is described in details Pressin
50. shold in per cent The new centre of the cluster is then recalculated This procedure especially improves the coherency of the cluster lattice when the clusters have irregular shape Below two cluster sets are presented originally found left side and recalculated after the refine cluster procedure right side 31 32 9 4 Example clustering parameter tuning Below a data set is presented with a strong difference in the unit cell vectors This is a case when careful tuning of clustering parameters is needed Parameters set radius 3 min points 20 Case Clusters along the reciprocal rods are not resolved Several rods are assigned to the same cluster Solution reduce min points to separate the rods Parameters set radius 3 min points 30 Case all rods are assigned to separate clusters now Clusters along the reciprocal rods are still not resolved Solution further reduce min points 32 33 Parameters set radius 3 min points 40 Case enlargement of the min points parameter did not cause any improvement The rods are clearly separated though Solution reduce radius Parameters set radius 2 min points 40 Case clusters along the rows start to separate the neighbourhood radius is still too large some clusters are fused Solution further reduce radius Parameters set radius 1 min points 40 Case clusters along the rows are separated it is possible now to define the
51. stallation ADT3D is platform independent software based on C and Qt It is available for Microsoft Windows XP SP3 Vista and Windows 7 For 3d viewing it uses the open source variant of the Coin3D and SIMVoleon scene graph library which is encapsulated in our open source viewer package To install ADT3D only two steps are necessary First install the free Gnuplot package which ADT3D enabled to generate many plots and second copying the ADT3D program files to your system 2 1 Gnuplot installation ADT3D will not run without a gnuplot installation on your computer system Therefore you have to download the actual version of gnuplot Version 4 2 patchlevel6 extract it install it where you want and add the binary folder to your system path Test the installation by opening a console command prompt and type wgnuplot exe If the gnuplot window appears gnuplot is ready to use gnuplot File Plot Expressions Functions General Axes Chart Styles 2D Help GHUPLOT Version 4 2 patchlevel 6 last modified Sep 2009 Syster H Hindous 32 bit Copyright C 1966 1993 1998 2004 2007 2009 Thonas Hillians Colin Kelley and nany others Tupe help to access the on line reference nahual The gnuplot FAQ is available fron http uuu gnuplot info t agi bend bug reports and suggestions to lt http sourceforge net projectssgnu plot erninal tupe set to windows giup lot 2 2 ADT3D installation 1 Create a new fold
52. t diffraction pattern presented as a surface plot based on the intensity values in each pixel A particular reflection can be selected by clicking on the diffraction pattern A bounding box is placed around the reflection A 3D plot of the reflection within the bounding box can be visualized by clicking the 3D plot bounding box button Intensity variation of the selected reflection through the tilt series can be visualized by pressing Rocking Curve function In the rocking curve intensities integrated within bounding boxes of different size are shown The largest box is 41x41 pixels the smallest is 5x5 The rocking plots are however more precise after some pre processing stages like pattern centering are done It is essential for rocking curves that the shift between the diffraction patterns is either minimal due to special precautions during the acquisition procedure or has already been corrected for Slice 0 3D plo pan Rocking Plot Position 520 154 Bounding box of Slice 1 Position 520 154 Intensity 12 13 7 2D functions 7 1 Data binning Most of the tasks do not require high resolution data as it is recorded The purpose of binning is to Save memory capacity while doing operations which do not require the full data volume Therefore for large diffraction patterns the binning Rebin MRC data icon is recommended at early stages of the processing and the corresponding function is included into the Standard functions
53. t digit 1 is considered as a counter and the rest of the name as a generic name It is assumed that all diffraction patterns have the same size in pixels the same diffraction acquisition parameters and the same orientation of the tilt axis diffraction rotation ak Zuletzt besucht Desktop EE p Bibliotheken A Computer pe a Netzwerk Me MDT _8 tif Me MDT_9 tif Me MDT_10 tif Me MDT_11 tif Me MDT_12 tif Shenin Bae SC mer Anderungsdatum 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 09 14 11 2011 17 17 14 11 2011 17 17 IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File IrfanView TIF File 8 196 KB 8 196 KB 8 196 KB 8 196 KB 8 196 KB 8 196 KB 8 196 KB 8 195 KB Me MDT_13 tif 14 11 2011 17 17 IrfanView TIF File 8 196 KB Ak nant 14 sit 14 141 0011 17 17 Ief 0 106 vo Views TIE Cila Dateiname m DT_1 tif X Dateityp Caf x After the first file has been selected the program asks for the generic name the name without the counter and extension Note that in this case the generic name includes the underscore symbol generic name MDT_ Most of the CCD cameras save the data in unsigned 16 bit tif format meaning th
54. tensity Gata SSL anserina aad eenden denderde diane 41 15 ADPENMEES arran a e e e a testen 45 13 1 ADT3D keyboard shortcuts and mouse keys un aanns eeens eeennseennnerennveennen 45 13 2 MRC file format as used in the software aans ssnnns seen ereensernnserennvsennen 46 159 CEUEAIe TOMA vonsrrtnenteetean een ed eee neet neee ne 51 15 4 Ditiraction pattern calibratiGn mers tekenenden eneen en wauetianess 52 1 Introduction The ADT3D software package is designed to process data collected from electron crystallographic experiments using the Automated Diffraction Tomography ADT technique Kolb et al Ultramicroscopy 2007 Kolb et al Ultramicroscopy 2008 Mugnaioli et al Ultramicroscopy 2009 The program has several basic functions First it is able to reconstruct a reciprocal space volume from a set of tilted diffraction pattern tilt series allowing the user to validate the quality of the data and thereby identify any complications as twinned crystals superstructures incommensurate structures or errors in data collection and reconstruction The program has an extensive set of functions to identify basis unit cell vectors Using the selected unit cell vectors it can index and integrate the intensities within the collected 3D reciprocal volume These intensities can then be used for structure solution using any structure analysis approach direct methods simulated annealing etc The ADT methodology brings a new level of improved d
55. the reciprocal vector basis is right handed i e the determinant of the vector matrix is positive When the matrix is no positive a warning appears in the log panel Log save cell fle canceled Logging Options Dialog accepted The cell is not right handed save cell fle canceled It is then the responcibility of the user to bring the cell to the correct settings 44 45 13 Appendices 13 1 ADT3D keyboard shortcuts and mouse keys The following table gives an overview on all available keyboard shortcuts and mouse interactions Ctrl O Ctrl S Ctrl Q Ctrl A Ctrl B Ctrl W Shift F4 Ctrl M Ctrl B Ctrl R Ctrl 2 Open the 2D viewer for the active dataset SSS en A rotate scene to view along cell vector 1 red rotate scene to view along cell vector 2 green C 1 B rotate scene to view along cell vector 3 blue 3D Viewer move mode gt gt gt 3D Viewer select mode __ 45 46 13 2 MRC file format as used in the software The MRC file is a file with extension mrc It contains a primary header and an extended header as described below followed by the data The file format is derived from the MRC file format which was originated at the Medical Research Council in Cambridge England The MRC format can contain a series of images in a single file The format consists of e A primary header A 1024 byte section e An extended header Additional information about the indiv
56. z az b1 bo b3 Ci Oe reciprocal space cell pixels deg la Ib Ie at BE y direct space cell A deg lal Ibl Ic apy 51 52 13 4 Diffraction pattern calibration It is generally assumed that once the camera length has been calibrated using a known standard material the pixel size in 1 nm can be calculated for every diffraction pattern acquired at this camera length This holds as long as the convergence of the electron beam used for diffraction experiment is either the same or is being accounted for In selected area diffraction experiments the electron beam is usually parallel enough to neglect the convergence effects However in nano electron diffraction mode the diameter of the beam fallen onto the sample is usually between 20 and 100 nm In this regime certain convergence of the beam is introduced and as the result diffraction patterns appear out of focus These patterns are then additionally focused using diffraction lens Additional focusing of diffraction patterns often causes rotation and expansion contraction of the whole pattern As a result the effective camera length and pixel size in 1 nm determined for the parallel beam conditions cannot be used for nano beam diffraction The error in dy measurement from a diffraction pattern acquired at strongly convergent conditions can be as large as 15 The value of pattern expansion contraction depends on the effective convergence of the beam It is rather
Download Pdf Manuals
Related Search