Analysis Tutorial Data

Keys, Commands & Abbreviations

Mouse functions

Keyboard shortcuts

Document Abbreviations

Loading Spectra

Start a new project

Start Analysis on the command line by typing:

-> analysis

Once the program has started and the main Analysis menu bar has appeared select M:Project:New. Into the query box that appears enter a name, without spaces, for the tutorial project, e.g. "TrialProject". Then click [OK] or press <Return>.

Now save the new project by selecting M:Project:Save. The first time you save a project the SaveAs dialog is brought up.  At this point you can choose to change the project name and/or the location where the project directory will be saved.  The default is that the project will be saved in the directory where you started Analysis.  The project directory is called the same name as your project (e.g. 'TrialProject').

After the first save, M:Project:Save will just automatically save to the existing location.  M:Project:SaveAs will let you save your project with another name, or in another location, but will still have the directory name the same as the project name.

It is a good idea to save now and again in case something goes wrong (either with your work or with the code). Save now by pressing [Save], without changing any of the information, and close the save dialog (you can use the red button at the bottom).

Opening Spectra from File

To open some spectra from file go to M:Experiment:Open Spectra. In the resultant Open Spectra popup select three spectrum parameter files to open. Do this by first navigating to the correct directory, where the downloaded tutorial data is saved (probably in a directory called "analysisTutorialData"). Once in the correct directory click on the line in top table corresponding to "5.spc.par", and then whilst holding down the <Shift> key click on the "7.spc.par" line. This will highlight three spectra to open.

If necessary enlarge the popup (click and drag the bottom edge down) so that you can see three entries in the bottom table. This table is a list of how the selected files will be placed into the CCPN data structure. Each file will be associated with an experiment and a spectrum that are in turn linked to a shift list. For the most part each experiment carries only one spectrum, as will be the case in this tutorial. However, it is possible to have one experiment with several spectra, e.g. when they are processed differently. All the spectra will use the same shift list, as the experimental conditions are the same, so leave that field alone.

Enter names for the three experiments by double clicking on the experiment name rows, e.g. "Expt_1", entering the name and then pressing <Return> or clicking outside the box. Set the names of the three experiments to "HSQC", "NOESY-HSQC" and "TOCSY-HSQC" respectively, or similar. The names of the spectra could also be altered, but this is not necessary here and the spectrum names being inherited from the spectrum file are informative.

Note that the file format option at the top op the popup is set to "Azara", which is correct for the files we will open. If the spectra were stored in a different format (Bruker, Felix, NMRPipe, NMRView or UCSF) this option would have to be altered. Finally click [Open Spectra].

Now, for each spectrum in turn, file and referencing verification dialogues will appear. Again, enlarge the popup so you can see the entire table. The {Verify Referencing} tab will be open - it shows that parameters you normally need to look at. The {Verify File Details} tab lets you edit where and how the file is stored, but this its not normally needed. Click [OK] on the button at the top of the popup (from either tab) to continue. You will get a warning about the data bit ordering of the file - just click [Yes] here. Note that you can change spectrum referencing and file setup at any time after loading via M:Experiment:Edit Spectra. For the second and third spectra you can toggle the "Is data big endian" option  in the {Verify File Details} tab to on before pressing [OK] - or you can continue as the first time.

Note that you may skip all the verification steps, if you know your parameters to be correct, by selecting the "Skip verification dialogs" option before you click [Open Spectra]. You can change almost all the spectrum referencing and file parameters at a later time should you need to make adjustments. The only thing that you cannot change after a spectrum is loaded are its primary axes; in terms of number and which isotopes they refer to.

Setting Experiment Types

After the verification there is one final popup where you specify the types of the experiments that were run. This information is very useful later on; for example Analysis uses the knowledge of which experimental dimensions correspond to single bond transfers to automatically remove impossible assignment options. For the HSQC spectrum set the category to "through-bond" and the synonym to "15N HSQC/HMQC" OR set the full type to "H[N]". Here the synonym is a human readable, but potentially ambiguous, name for the type of experiment whereas the full type is a precise, but often less readable, descriptor.

Set the synonym for the second experiment to "15N NOESY-HSQC" and the third experiment to "15N TOCSY-HSQC". Note that for these experiments there are two different full types with the same synonym (double click the Full Type column). For example the "15N NOESY-HSQC" could be either H_H[N].NOESY  or H[N]_H.NOESY, with the difference being that the NOESY transfer can come either  before or after the HSQC step. For our experiments the selections H_H[N].NOESY and H_H[N].TOCSY are correct. When the experiment types are fully set click [Done]. You may have to expand the popup to see the [Done] button. You can now also close the Open Spectra popup.

Spectrum Windows and Peaks

Window Components and Views

Once the three spectra have loaded, two windows will automatically be created. One is a two-dimensional window "window1" containing the HSQC and the other "window2" is three-dimensional for the NOESY and TOCSY. Note that if you minimise a window you can get it back again by going to M:Windows:HN: window1 (or equivalent).

You can move around the spectra within the windows using several different inputs. To zoom in and out use the <PgUp> and <PgDn> keys, the middle mouse wheel (if you have one), or hold <Shift>, click the middle mouse button, and move the mouse up and down.  If you zoom out from your spectra so that you can see their edge you will notice a dotted line which denotes their border.

You cannot zoom out further than the maximum size allowed in each dimension.  To change this size open up M:Window:Axes and select the Axis Types tab (if it is not already selected).  Each Axis Type has a Region, which can be edited by double clicking on the relevant cell.

To pan around the spectra you can click and drag the scrollbars at the edges of the windows, use the arrow keys, or click and drag on the spectra with the middle mouse button.

Note that you can change the way in which the arrow keys move the spectra but going to M:Other:General Options and select the {Main Profiles} tab if it is not already selected.  In the row with Parameter "Keys Pan Window View", setting the corresponding Value to True means that the keys act as if to move the view, rather than the background.

Whilst in "window2" move the extra scrollbar at the very bottom of the window. - Left click and drag. This scrollbar is present on the 3D window to change the depth or plane of the spectra being viewed. On a 4D window there would be yet another scrollbar. To change the thickness of the displayed planes click and drag the side of the depth slider with the middle mouse button.

To move to a specific coordinate in depth (so z) dimension you can type a location into the "15N" box (and press <Return>) at the bottom left of the window. To center the screen (so xy) axes at a given position you place the cursor at the location and type 'c' (if the window is in focus) or click on the coordinates button at the top bar of the window (to the right of "Strips") to get a dialog box, where you can type in an explicit position.

Staying in window2, click on "Spectra" at the top of the window. Here you will see two colored buttons, one for each of the 3D spectra. If you click the buttons you can independently toggle the contour displays for the two spectra on and off.

Window Options

There are various parameters that you can manipulate to change the way in which windows and spectra are displayed. Firstly we will make a new window which we can play with, whilst leaving the others alone. So go to M:Windows:New Window. Select "1H" in the z1 pulldown menu to make a 3D window which is orthogonal to the current window. Note that such a window is capable of displaying all three spectra because the X-Y (screen) plane has 1H and 15N axes and this is valid for the HSQC as will as the 3D spectra. Toggle some of the spectra to visible by double clicking in the second column of the table and then click [Create].

We will manipulate this window via M:Windows:Edit Windows, in the {Window & Axes} tab. For the "window3" row of the top table try changing the aspect ratio by double clicking on the 'Aspect Ratio' column. Also try adding 'Crosshair Traces' (top table) or 'Side Traces' (lower table) to the axes. Note that if you really don't like the scrollbars in the window you can turn them off here.

Slice panels (1-dimensional cross sections) can also be added to and removed from the windows via the mouse-menu (accessed with right click over the window), selecting the options in R:Window. To change the intensity scale of the spectra in the slice panels use the <Home> and <End> keys

Setting Contours

Now we will investigate changing the contours. Firstly change the contour colours by going to M:Experiments:Edit Spectra, selecting the {Display Options} tab and double-clicking in the "Positive colors" column. To set the contour levels, within a spectrum window go to the top "Contour" option. The green arrows will raise or lower the contouring floor, while the +1/-1 will change the number of contours.

To alter more of the contour settings now click on [More].  In the popup that appears, select an appropriate spectrum and set the levels as desired. You can quickly adjust the bottom most or "Base level" contour with the "/2" (divide by two) and "*2" (multiply by two) buttons. Note that any changes you make in the "Auto contour levels" panel will create the specified number of levels in a geometric series (e.g. 1, 2, 4, 8, ...), starting at the base level using the specified multiplier. If the "Add levels" option is on, the levels will be created by the addition of an increment, rather than by multiplication.

To add negative contours, set the "Negative" option on. If you wish to have a only one negative contour, for example to give an indication of truncation artifacts without negative contours generally, then select (with a mouse click and drag) the unwanted contour values in the "Negative levels" panel and press <Del> then click [Apply Manual Edits] or press <Return>. When you are finished adjusting the contour levels click [Done] to close the popup.

Peak Picking and Manipulation

Focusing on the HSQC spectrum the next task is to define some contour extrema as peaks. There are two common ways to pick peaks. One is to search for all extrema in a boxed region. Try this in the HSQC window by holding down <Shift> and <Ctrl> whilst clicking with the left mouse button and dragging the box to define a rectangular pick region.

Note that when the peaks are picked the contours will carry diagonal crosses to denote the peak position.  Each peak will also have some annotation to the top right of the cross, which we discuss below.

Locate the contours near the point at 8.7 ppm on the 1H axis and 118.4 on the 15N axis.  These contours represent two overlapping peaks where the extrema search will only pick one of the peaks. To define the second peak position hold down <Ctrl> and click on the location to pick the new peak.

Now we will select some of the peaks, e.g. artifacts or noise, for deletion. To delete peaks click with the left mouse button and drag a box over a region containing peak crosses (without holding down any keys). When the mouse button is released you will see that the peaks in the defined region are highlighted with a border around the cross. To select just a single peak click near its centre with the left mouse button. To delete the selected peaks press <Del> or select R:Peak:Delete selected peaks.

If you have peaks selected and then select a different set of peaks you will see that the selection is completely substituted for a new one. You can add to an existing peak selection by holding <Shift> while you choose. Note that you can select peaks in several different spectra, and from different windows, in this manner.

Next try moving a peak. For this select a single peak, move the mouse cursor to the desired position and press the <p> key. Alternatively you can hold the <p> while you move the mouse. If the selected peak is near, but not exactly at, an extremum you can shift it to the extremum with <P> (i.e. <Shift> + <p>), as long as there is no existing peak already at that position (in the same peak list).

To open a table containing a list of peaks, select M:Peaks:Peak Lists. In the {Peak Table} tab select a spectrum for which you have picked peaks with the top left pulldown menu. Click on one of the peak rows so that it is highlighted, then click on [Find Peak] - remember to select the correct window with the pulldown to the left of the command button. This will locate the peak and centre it within its window.

With the table and the spectrum window both visible, select the peak in the contour window and move it to a different position. - Watch in the peak list table as the peaks ppm values are continuously updated. This is a general feature of tables in Analysis. The tables are updated when things change, to reflect the immediate state of the underlying data.

It is also possible to move a peak by editing its position value in the table. Do this by double clicking on a "Shift" column in the table for one of the peaks that you can see in the spectrum window. Edit the peak position for that dimension, but not so much that the peak would not be visible on screen, and press <Return> to see the peak move relative to the contours. 

Markers and Navigation

Pick and select an isolated peak in the HSQC spectrum. Put a mark through it by holding the cursor over the peak centre and pressing <m>. The lines produced are a multi-dimensional marker at the peak position and will be visible at the equivalent 1H-15N location in the 3D window. To go to this equivalent position in window2, with the cursor over the marked peak select R:Navigate:1H - 15N in window 2. Note that there were two navigation options, where the second option would take you to the 1H position on the vertical axis, rather than the horizontal axis, which represents the amide proton.

Multi-dimensional marks, vertical ruler lines and horizontal ruler lines can be added to any window location, not just on peaks, using the <m>, <v> and <h> keys respectively. To increase the number of marks and rulers that can be displayed at one time select M:Window:Marks and Rulers or R:Markers:Options. Note that you can clear all marks and rulers with the <n> key.

Strips and Strip Navigation

Now we will start to manipulate "strips" which are sub-divisions of a window that are connected (in terms of their view) in one of the screen dimensions, but independent in the other dimensions. Go to window2, select the "Strips" option at the top and click [+], this will add a vertical division to the window. Click and drag with the middle mouse button to move the spectra - you will see that the vertical axes of the strips are tied together, but the horizontal axis is independent. The depth dimensions are also independent, e.g. if you move the bottom most, 15N scrollbar. Which depth dimensions are moved depends upon which strip is active. The active strip is indicated by an asterisk "*" next to its strip number and is set either by double-clicking (left mouse) within a strip or by using the strip options at the top of the window; by clicking [1] or [2] or whatever. The active strip is also the one that will be removed when the [-] button is clicked. You can rearrange the strip order, moving the selected strip with the green arrows, and swap between vertical and horizontal strips with the toggle button. To remove all the strips press [Clear].

Manually locating strips at interesting positions can be tedious, but there are various options to build strips from pre-defined locations, for example using peaks. To make strips using peak locations select three (picked) peaks in the HSQC spectrum and select R:Strip:Peak Location Strips:1H-15N in window2 (the same option is found in R:Navigate). The result is three strips located at the amide positions corresponding to the HSQC peaks.

Assigning Peaks to Resonances and Spin Systems

Assigning New Resonances

Given that we now have some peaks and can move between the different spectra we will start some assignment of the peaks to resonances. Initially we will assign the peaks anonymously, that is to say we will link peaks to a resonance number, but not say which atoms the resonances comes from. Initially such an assignment is not very useful, but we can go on to link related peaks to the same resonance numbers. For example we can say that a whole column of peaks in the 3D TOCSY and NOESY and an HSQC peak are derived from the same amide resonances. When we specify which atoms the resonances derive from then all of these linked peaks will automatically be assigned to those atoms.

To start an assignment choose the isolated HSQC peak at  the location 7.12, 119.51 (1H,15N) , and with the cursor near its centre press <a> (or you can select R:Assign:Assign HQSC...). Note it doesn't matter which peak you choose really, but we will be referring to this one later. The Edit Assignment popup will appear containing two rows of tables, one for each of the HSQC dimensions. In the left most table of each row click the button [<New>], this will add the resonances [1] and [2] to the 1H and 15N dimensions of the peak. Now click [Set Same Spin System] at the bottom left of the popup. The resonance assignments will become {1}[1] and {1}[2] here the {1} annotation signifies that the resonances are both in spin system number 1, which in this instance indicates that they both belong in the same amino acid residue.

Now mark the peak (<m>) and navigate to the equivalent position in TOCSY and NOESY spectra in window2 via R:Navigate. Ensure the peaks along the marker in window2 are picked (<Shift> + <Ctrl> + left click and drag). Note that if you cannot pick some 3D peaks they may not have a maximum within the selected depth range. If this the case you can adjust the depth position or width.

Now assign one of the 3D peaks at the marked amide location: Press <a> with the cursor over the peak and find the Edit Assignment popup. You will see that the popup has now updated for the 3D peak and consequently there are three dimension rows. Because the peak position closely matches the chemical shift value for resonances [1] and [2] they appear in the right hand tables. We can link these existing resonances to the 3D peak by clicking on their rows in the right-hand table. When you do this the resonance annotation appears on the left hand side (and in the spectrum window) to indicate that the peak is now linked, i.e. assigned to the resonances. Note that you can remove the resonance assignment in the popup by selecting a row from the left hand side and clicking [Clear contrib]. The use of the term "contrib" here reflects the fact that a given peak dimension could potentially have a contribution from multiple resonances. So for example you can could have an ambiguously assigned NOESY peak where two different pairs of close resonances contribute to the measured peak intensity.

With the 3D peak's amide dimensions fully linked we will quickly give the other 3D peaks at the same amide position the same assignment. Do this by zooming out in the window so that you can see all the peaks along the marker line, select all the peaks including the assigned one (left click & drag) then select R:Assign:Propagate assignments. This will cause the resonances assignments {1}[1] & {1}[2] (displayed on the spectra as "{1}[1],-,[2]") to be spread appropriately across all the selected peaks.

Efficient Resonance Assignment

The linking together of related peaks in different spectra by assigning them to common (anonymous) resonances is something that can be partially automated to speed up the assignment process. We can use the HSQC positions to define unique amide locations and pick and assign related spectra base upon these "root" locations. The first step in this automation is to define new amide resonance and spin system identifiers for all the peaks within the HSQC spectrum. Initially ensure you have most of the HSQC peaks picked. Then select M:Assignment:Initialise Roots.

When you see the 'Initialise Peak Lists' popup, there is a table called 'Amide Sidechain Peaks' with a few rows filled. Some of the peaks which you picked in the HSQC will be from NH2 groups of amine side chains, and you need to handle those before you can initialise the peak list. Clearly the NH2 groups give two peaks, one for each hydrogen but both have the same 15N resonance (and thus 15N chemical shift). If such pairs of peaks were processed in the same manner as the backbone amide peaks they would become linked to two different pairs of resonances in two spin systems, when in reality they should carry the same 15N resonance and be in only one (side chain) spin system. Click on a row of the table to view the peaks, first making sure you set it to follow the right window (here window 1 or 3). If you think this looks like side chain NH peaks, double click the 'Confirmed' column so it changes to 'Yes'. When you are happy with all of them, click 'Initialise Peak List!' at the bottom.

This command calls the resonance and spin system assignment routines, that we used to assign the first peak, on all of the HSQC peaks. Note that the routine knows which spectrum to work with because we set the experiment type of the HSQC correctly as H[N]. You will now see that all of the peaks carry assignments of the form {x}[y],[z]. If you look at the NH2 peaks that you confirmed, you will see that both peaks belong to the same spin system and that the 15N dimension is assigned to the same Resonance.

The next part of the assignment process is to link resonances, as we did before, from the HSQC to the corresponding trains of peaks in the 3D experiments. From the menu select M:Assignment:Link Peak Lists. In the Link Peak Lists popup that appears ensure that window1 is selected in the Root Window pulldown menu and select window2 in the pulldown menu at the bottom right, and click [Add Window:]. Now take a quick look at the {Spectra & Tolerances} tab, check if the parameters look OK, and go on to the {Link Peaks} tab. You will notice that the peaks from the HSQC are listed in the table. If you click on one of the rows, window1 will centre on that peak and the location of window2 will move to the same amide frequencies. Rearrange the positions of window1 and window2 so that you can clearly see both and the popup. Select a row corresponding to an HSQC peak that is not overlapped and click [Pick & Assign Root Resonances]. You will see that peaks are picked in the 3D spectra, in the box defined by the tolerances, and assigned to the amide resonances from the HSQC. You can go on to further amide positions by clicking [Next Root]. With appropriately set tolerances you may also click [Pick All & Assign Root Resonances] to process all of the amide positions - you can still use the 'Next Root' function (etc.) to loop through the peaks afterwards. Note that closely overlapping amide resonances would still have to be checked or linked by hand.

In the {Spectra & Tolerances} tab, on the row corresponding to the TOCSY, double click on the "Assign Non-root dim?" column. Then after having linked one of the amide spin systems to the 3D experiments, click [Assign Target Non-root Resonances]. You will see that the F2 dimension of the TOCSY peaks are now each assigned to a different resonance (which we could then go on to assign to the appropriate NOESY peak).

It is important when picking peaks and assigning resonances in this automated manner that noise and artifact peaks are not picked. Of course any offending peaks can be deleted afterwards, but most can be avoided by setting the picking tolerances to appropriately small values and setting the contour levels so that the noise is not visible. By default the peaks are picked only above the visible contour base level.

Now that we have defined and linked many resonances, look in the main resonance table at M:Resonance:Resonance Table. You will see that all of the resonances are listed here and many operations can be performed on them. The important thing to note here is that the chemical shift of each resonance is automatically calculated from the positions of the peaks to which it is assigned. Note that a resonance assigned to only one peak will have no deviation in its shift, but those assigned to several 3D peaks will deviate as the amide peak position varies slightly. By default the chemical shift values are the average of the assigned peak positions, where each spectrum is weighted equally. However, different dimensions of different spectra can carry different shift weightings (set at M:Experiment:Edit Spectra {Tolerances}) so that the value of a shift may be influenced more by the more precise experiments.

Defining Sequences and Molecular Systems

So far we have been linking related spectra with anonymous resonances. In the end we will assign these resonances to specific atoms, but first we must specify the molecules within our experimental sample that we may assign to. Creating a list of atoms to which we may assign has two distinct steps. The first is the specification of a molecule, usually with an amino acid sequence, that will act as a template. The second step is to build an assignable molecular system from the templates to give the atoms. The molecular system represents all the assignable molecules that may reside within the NMR sample, these include proteins, nucleic acids and small molecules. For this tutorial we make a mock molecular system with one protein chain and one small molecule. These two steps exist so that we may define a template sequence once, but have the potential to create several distinct polypeptide chains for assignment, for example if we have a homodimer.

Open M:Molecules:Add Sequence. This will open the 'Edit Molecular Information' popup on the {Add Sequence} tab. 'Add Sequence' is a shortcut command that can create both the molecule and molecular system and add a new sequence to it, all in a single operation (it can also add to existing molecules / molecular systems). The kind of molecule you want to create is controlled by the switches at the top. First set 'Input Type' to 3-Letter/Ccp. Then cut and paste the following sequence (left mouse to select in the browser and middle mouse to paste) into the text window:

LYS ALA SER SER PRO SER SER LEU THR TYR LYS GLU MET ILE LEU LYS SER MET PRO GLN 
LEU ASN ASP GLY LYS GLY SER SER ARG ILE VAL LEU LYS LYS TYR VAL LYS ASP THR TYR   

The input sequence does not have to be perfectly formatted. You can click [Tidy] at any time to see how Analysis has interpreted the sequence. Also, you can switch between one- and three-letter codes after the sequences has been entered. If you look in the Ccp Codes pull-down menu (top right) you will see all of the residue codes that are currently available, many of which are modified amino acids. Now set both 'Destination Molecule' and 'Destination Mol System' to <New>, and click 'Add Sequence!'. You will be prompted for the names of the new molecular system, and chain. When finished the popup automatically switches over to the {Chains} tab, where you can see the result.

Just for fun, we will now add a small molecule to our project. This could be part of the first molecule we made, i.e. linked to the protein in some way, but instead we will make a new molecule. 'Add Sequence' is only relevant for linear polymers so we shall do it another way. First go to the {Small Compounds} tab. Begin by choosing [DNA] in the 'Mol Type' pulldown at the top then select a compound in the right hand table to display an idealised structure in the left hand panel. You can rotate the cartoon compound by clicking and dragging with the middle mouse button. Using <Shift> or mouse wheel while dragging up and down with the middle mouse button will change the zoom level.
With any compound selected set 'Destination Molecule' to '<New>' and click [Add Compound] to enter it into your second molecule template - this will bring you to the {Sequences} tab. Go back to the {Small Compounds} tab and now try adding a non-polymer molecule. Set the 'Mol Type' pulldown to 'Other' and select a molecule at random. Some simple molecules (like Zn) are available, but most will give you a popup asking if it is OK to download the molecule description. If you say 'Yes' the description will be downloaded (assuming you have an internet connection and write access to the relevant directory), but even if successful most likely you will also get a warning that there are no coordinates available (that is why we started with DNA). So select something like ATP or Zn.  Add that compound to another new molecule.

To make molecular system information from your templates, go to the {Chains} tab. Set 'Mol System for new chain' to '<New>', select one of your small molecules as the template, and click [Make Chain From Template]. Then set a new template, keeping the setting for the Mol System, and click [Make Chain From Template] again. You are done creating molecules now; try to look in the various tabs and see what information is available.

To view the new polymer chains have a look at M:Molecules:Atom Browser. You can scroll through the sequence and by toggling the [H], [N] etc. you can display all of the assignable sets of atoms.

Assigning Resonances to Atoms

Once we possess sufficient information, the resonances to which peaks are linked may be assigned to specific atoms. For the purposes of this tutorial we will cheat with prior knowledge and link spin system number 1 "{1}" to Leucine 21. Find the HSQC peak that we first assigned by going to M:Peaks:Peak List {Peak Table} and selecting the HQSC from the pulldown menu. Click on the row of the {1}[1] {1}[2] peak and click [Find Peak]. (If it is difficult to find then click on the "Assign. F1" column heading which will sort that column.)  In reality we could choose any peak that we have assigned to the spin system and connect its resonances to 21 Leu Hn and 21 Leu N.

With the cursor over the centre of the peak press <a>, (or select R:Assign:Assign peak from the right mouse menu). Once the popup appears select the first hydrogen resonance (top-left table) and then click on [Assign {1}[1]]. This will bring up the Browse atoms popup. Make sure the correct Chain is selected in the pulldown menu, and make sure that the hydrogen and nitrogen atoms are displayed (click [H] and [N]), then scroll down to click on H in the 21 Leu column. Return to the Edit Assignment popup and you will see that the resonance is now annotated as 21LeuH and that the 15N resonance has been assigned (note that we use the IUPAC convention of naming amide hydrogens in proteins as "H" rather than "Hn"). The 15N resonance was assigned because it is directly bound to the 1H resonance - Analysis knows this because we have set the experiment type to H[N]. Return to the Atom Browser and you will note that the atom sets you have chosen in the browser will change colour to reflect their assigned status.
Now that the Resonances are connected to atoms you will see that the peak annotation has changed to "21LeuH,N", not only for the peak we directly assigned, but also for all of the 3D peaks, that are linked to the same resonances. If you look in M:Resonances:Spin systems {Assignments} you will also see that the spin system has been unambiguously assigned to the leucine residue by virtue of our choice of atoms.

During such atomic assignments it is may be useful to have representative chemical shift values for the possible atomic assignments. These can be accessed at any time at M:Resonance:Reference Chemical Shifts by selecting the appropriate residue code and atom type (heavy atoms or hydrogens).

Next we will make an inter-residue connection. In window2 mark (<m>) the NOESY 21 Leu derived peak just below the diagonal at 7.4 ppm. Returning to window1 you will note that two vertical lines are present on the HSQC, one through the 21 Leu spin system and the other at the amide H shift of a nearby residue. Looking at the HSQC along the line at 7.4 ppm we see that two peaks are intersected, although neither is dead centre (the marked NOESY peak is distorted by the diagonal peak). Cheating again, the peak at 111 ppm in the 15N dimension is the residue N-terminal to 21 Leu.

We will bring the 3D TOCSY and NOESY peaks of this neighbouring spin system up in a new window strip: In the 2D windows select both the 21 Leu HSQC peak and the peak at 7.36, 111.0 ppm by (left click and drag with <Shift>) so that they are highlighted by surrounding squares. Now in the right mouse menu select R:Strip:Peak Location Strips:1H - 15N in window2. Alternatively we could have manually added a new strip to window2 and, after having activated the destination strip, used the same navigation function as before in the second HSQC peak.

In the new strip of window2 you will see the neighbouring spin system with its return amide-amide NOESY peaks. The spin system in this strip is 20 Gln. So as before, pick the appropriate peaks in the spin system, assign one of them to the resonances in the HSQC spectrum, then propagate the assignments to the remainder.

Finally we will assign the connecting amide-amide crosspeaks. In the 21 Leu strip select R:Assign:Assign peak over the marked peak (just below the diagonal). In the assignment popup you should select the 20 Gln H resonance for the second hydrogen dimension. Now assign the reciprocal amide-amide NOESY peak in second strip. With the assignment complete, select M:Charts:Assignment Graph and then in the {Options} tab click on the button named after your 3D NOESY experiment. Returning to the {Graph} tab you will then see the connection you have just made to the spin systems in the graphical display. 

Loading Existing Projects and Save Options

Quit the current Analysis session, saving the extant project as you see fit.  Next we look at a pre-existing project.

Now at the Linux/Mac shell command line type:

-> ls 

And press <Return>. You will see that we have a directory called TutorialProject in your current directory. This directory contains all the user data for the project that we are about to open. The data are spread over many directories and files. It is worth noting that the various XML files, which we shall see in a moment, are only loaded when required, when we access the data. In this way Analysis only loads the files it needs. Conversely, when a project is saved, only the files corresponding to modified data packages will be written. The application will automatically search its data directories for files, so the data available will mirror exactly what is in the directories.

Assuming you are starting inside the directory containing the downloaded tutorial data. At the command line now type:

  -> tree TutorialProject/

Or if the "tree" command is not available:

  -> ls -R TutorialProject/

This command will list all of the contents of the project directory. You will see that there are several layers of directories with XML files within the directory structure. Each bottom directory contains files with a different type of data that may be loaded into the project. The rather cryptic file names are generated automatically, using the key attributes of the top object in each file, the project and user name used to generate them, and a timestamp that serves to make a globally unique identifier. The names are used by the underlying CCPN implementation and can not be renamed by hand. Reference data distributed with the release (chemical compound templates, reference chemical shifts etc.) live in another location, as do user-specific profile data. Advanced users can control which data live in which directories. Each project has a file that contains all the bookkeeping information to specify which data live where; it is in the TutorialProject/memops/Implementation directory. When you copy a project to a different location, the internal file references will no longer be pointing to the right on-disk locations. In such circumstance Analysis will look for the new file locations and upgrade the directory paths automatically - as long as you are using the standard directory setup this should not be problematical.

When the CCPN XML file formats are upgraded to a new version, the new code will automatically be able to read the old formats. For rare cases where this is not possible, like the big jump between CCPN version 1 and version 2, there is an upgrade server available.

Start Analysis with the existing project by typing:

  -> analysis TutorialProject

The project will take a few moments to load, and it might tell you that the directories have moved. Also, the locations of the spectrum (binary) data will most likely have changed. If the "Edit Project Spectrum & Data Paths" popup appears, double click on a red row in the "Absolute Path" column and in the file dialog navigate to the "TutorialProject/spectra" directory. When inside the "spectra" directory, click [OK] and you will hopefully note that the red colour has gone, indicating that the spectrum files have been located. Close the popup and when it is done loading several spectrum windows will appear on screen.

M:Project:Save As allows you to save the user data in a new location. M:Project:Backup allows you to make backups, and to set the auto-backup options. There is a third place to control file locations: M:Experiment:Edit Spectra:{File Details} controls the storage location of spectrum data files. It is anticipated that your binary spectrum files will usually residue at a location somewhat separate from the CCPN XML files.

The file locations for binary spectrum data cannot be upgraded automatically and may well get out of sync when you move projects. If a spectrum file is seemingly missing at the location specified by the CCPN project then Analysis gives you the opportunity to re-point the spectrum to the correct on-disk location. Often this can be done efficiently, when spectra are all housed in the same directory, by swapping the common part of the spectrum locations in a single step (i.e. as described above).

Also, at any time should you wish to change a spectrum location, or even move the CCPN link to a differently processed binary file, you can type in the file location of a single spectrum via M:Edit Spectra: {File Details} - File Name or repoint en masse by swapping directories at M:Edit Spectra:{Data Locations}.

Working With Variable Chemical Shifts

So far during this tutorial all of the chemical shift values we have been recording, by picking peaks and linking resonances & atoms, have gone into one single list. However, there are several circumstances when it is helpful to use more than one list of chemical shifts, so that you may for example separate sets of shifts based upon the experimental conditions they relate to; you may use different shift lists for different temperatures or when doing titration experiments. Commonly in such circumstances the underlying resonances (and hence atoms) that we work with are the same, but the positions of the peaks move significantly and we would not want to average chemical shift values over such a variation.

We are going to set up a new shift list for the second HSQC experiment present in the demonstration project. This second HSQC experiment relates to exactly the same sample (molecule) as the first HSQC but was recorded at a different temperature, which has caused the peak positions to move. Given that the second HSQC experiment is already loaded, all we need to do is go to the M:Experiment:Edit Experiments {Experiments} and in the Shift List column double click on the cell corresponding to HSQC_II. In the pulldown menu that appears simply select <New>. You will see that a new shift list is now listed. From now on whenever an assignment is made in HSQC_II the shift remains separate from the other spectra. Indeed, any resonances that were already connected to HSQC_II before its shift list was changed will have their shifts recalculated in two separate groups. You could move the experiment back to the original shift list at any time and the chemical shift values will be appropriately recalculated. However, if you disconnect experiments from a shift list (or delete a peak etc) you might have chemical shift values that are not defined by any peaks. Under such circumstances the chemical shifts persist, but are described as "orphans".

Assignment of HSQC_II, with its separately curated shifts proceeds pretty much as assignment normally does, with the only major difference being that for a new shift list (i.e. without any assignments under it) there will be no resonance possibilities that can appear when assigning a peak. - The only known shifts are distinctly different, representing different conditions.

To assign HSQC_II, even though we don't have shifts set, we can say that certain peaks in the two HSQC spectra are equivalent if we can see how they have moved under the different situations. Accordingly we can select a peak in the first HSQC and the corresponding peak in HSQC_II and propagate assignments. That is to say that the peak dimensions link to the same resonances (and hence atoms), even though things have moved. Go to spectrum window 1 and try this for the two peaks around 8.33,114.5 ppm (so one peak in each of the HSQC spectra). Use the mouse with a left click to ensure that these peaks (and only these peaks) are selected. Then in the right mouse menu select R:Assign:Propagate Assignments. Note that this function usually checks to ensure resonance positions are within tolerances, but here with the two shift lists it cannot so the connection is made anyway, which is exactly what we want.

You may like to repeat the above procedure for several HSQC peaks and their counterparts. With a few resonances represented in the new shift list (should be shiftlist 2) have a look at the resonance table M:Resonance:Resonance Table to see evidence of these links. In the resonance table note that you can choose between the different shift lists in a pulldown menu at the top, i.e set the ShiftList: pulldown to "ShiftList 2". Note that when the shift list changes the resonance positions shift and the number of peaks linked also changes. Apart from the resonances, Analysis also provides direct access to the shift measurements. Looking in the menu at M:Data Analysis:Measurement Lists, you will see that there are tables of the shift values corresponding to each list. Note that other types of value like T1 rates, or Hetero-NOE values will also appear as measurements in these tables if they are calculated.

Copying peaks between related spectra

Propagating assignments between peaks by manually selecting them, although sometimes necessary, can be tedious if you have large number of peaks that are equivalent and have closely matching positions. In this instance Analysis provides a utility to compare all of the peak positions in two related spectra and transfer the resonance assignments efficiently, whilst still leaving sufficient scope for human intervention. 

In this exercise we will transfer the amide assignments from the peaks in the HSQC experiment to the HSQC_II experiment. In the main menu we select M:Assignment:Copy Assignments. In the Copy Assignments popup ensure that the first "HQSC" experiment is set as the Source Peak List and that the second "HQSC_II" is set as the Target Peak List.  In the top table you will see a list of the source peaks, together with an indication of how many destination peaks they match, within the given tolerances. Note that when the Follow Peaks? option is set clicking on a source row will cause peak to be highlighted in a relevant window (window1). Also, selecting the source will display the potentially matching destination peaks in the lower table. Clicking on a target row will, as before, highlight the relevant peak.

This utility can be used in one of two ways. One can manually choose the correct destination for each source peak and [Assign Selected Target], or if you trust the completeness and accuracy of the destination one can automatically assign the closest destination peak to each source one with [Assign Singly Matched]. When taking the more automated approach it is common to start with strict tolerances to assign the more certain matches and then relax the tolerances to fill in the remainder.

Sequential Protein Backbone Assignments

We will now look at the sequentially assigning protein backbone spin systems using triple-resonance experiments. There are two basic, although linked, parts to the process. The first is the linking of sequential spin systems (collections of resonances that relate to one residue) on the basis of matching peak positions. The second is the matching of runs of unassigned spin systems to residues within a sequence.

Spectrum Setup

Select M:Assignment:Link Seq Spin Systems. Now make sure that window2 is visible and expanded along one side of the screen. With the mouse over window2, select R:Window:Clone. This will produce an identical window containing the triple resonance backbone spectra. Now arrange window2 and the new window5 so that they are side by side, with window2 shrunk to about one quarter the width of window5 (both windows should be as tall as possible). Then switch off the HNCA spectrum in window2, via the "Spectrum" tab, so that only the HNcoCA is visible.

In the Link Sequential Spin Systems popup select window2 as the "Query" window and window5 as the "Matches" window. Now click on the [HNcoCA:111] button in the top "Query" panel and the [HNCA:110] button in the "Matches" panel. This setup means that we are going to compare specified 13C peak positions in the HNcoCA experiment with potentially matching peak positions in the HNCA experiment.

The rational here is that the HNcoCA experiment's peaks have the carbon shift of the preceding alpha carbon along the polypeptide chain at a given amide location, where the HNCA has both the intra-residue and preceding alpha carbon peaks for each amide. Thus we can potentially use both spectra to say two amide spin systems are sequentially connected by saying that an inter-residue peak of the HNcoCA derives from the same resonance as an intra-residue peak of the HNCA.

Note that this system can readily use other backbone experiments like HNcoCACB, HNCO, HB/HAcoNH etc. with the same approach and that this tutorial only uses the alpha carbon experiments for simplicity.

Linking Sequential Spin Systems

Now go to the {Spin System Table} tab and click on the row that corresponds to spin system 90. You will see that the two triple resonance windows move to new locations. The location of window2 is at spin system {90} as found in the HNcoCA. The window with the HNCA, window5, has moved to the position of any peaks that match the alpha carbon frequency. In this case there is only one good match, which corresponds to spin system 89. Note that if the "Filter by Inter/Intra Type" option in the {Settings} tab is set to off then spin system {90} matches two HNCA peaks, but we know that the spin system {89} peak is the correct one because the 76Lys peak is already assigned and because the 76Lys HNCA peak is not an intra-residue one, i.e. it is coincident with the 76Lys HNcoCA peak. Click to highlight the row corresponding to spin system {89} in the matched peaks panel, and then click [Set Seq Link]. You will now see that the tables of the popup update to show that {90} is set as "i+1" of {89}. Also note that in the spectrum window the peak annotations of the aligned  HNcoCA and HNCA peaks have changed to illustrate that they are both assigned to the same 13C resonance. Now click [Goto i-1], which will repeat the carbon shift matching, but this time from a sequence position one earlier in the sequence.  For spin system {89} apply [Set Seq Link] for the only match to{171}, then [Goto i-1] again. Repeat the sequential linking procedure for {171} & {161}.

Assigning Segments to the Sequence

You will see that as you select the various spin systems in the main table, in the lower right Residue Types table there is a display of the probable types of amino acid residue. This prediction is based upon how well the shifts within the spin system match the chemical shifts in the BMRB database. As spin systems are connected sequentially, amino acid type predictions are made for the whole sequentially connected section. In the lower left hand table the connected spin systems, given their probably amino acid types, are matched to the protein sequence. Here the highest scoring positions of residue type match for various five residue sections are listed. You will see that the assigned residue positions are coloured blue, but that there is a grey section (starting st Ser 8) which is not assigned that matches the section we have just linked with high probability. Simply select the row for this unassigned selection and click [Assign Selected]. You will see that all the residues in the section (by virtue of their links for the most part) become assigned to the selected section.

Select the option M:Moleules: Atom Browser.  Make sure that the element [H] is displayed (click the button to get the green hydrogen assignment options) and select the "H" (amide) atom for 8 Ser. You will see that not only is 8Ser assigned (i.e. the atom option goes dark green), but the atoms in the residues which we just connected sequentially are also assigned. Look through the spectra and the Link Sequential Spin Systems to verify that all the connected spin systems have been assigned. Note that in this instance it was possible to assign the resonances to unique atoms, as well as assign the backbone spin systems to the sequence, because the resonances had their atom type set previously.

Structures and NOEs

The penultimate part of the exercise is to look at how we can use Analysis to generate distance restraints that may be used in a structure calculation and how we can pass intermediate structural information back into Analysis to help with NOE peak assignment.

Distance Restraint Lists

To make a list of distance constraints from the assigned peaks in an NOE peak list first go to M:Structure:Make Distance Restraints. At the top of the popup, change the peak list to N-NOESY:182:1. We can leave all of the other parameters alone for demonstration purposes. The Restraint Distance Params section would allow us to specify how the NOE peak intensities relate to the distance bounds of any generated distance constraints. The default method is to calculate a target distance as peak volume raised to the power of -1/6 multiplied by some scaling factor, such that the reference intensity (in this case defaults to the peak list's average volume) exactly corresponds to the reference distance (in this case 3.2 Angstroms). The upper and lower bounds of the distance constraint are calculated as fractional changes from the calculated target distance (default is 20% above and below) while observing absolute minimum and maximum values for the bounds (1.72 & 8.00 Angstroms respectively by default). The {Residue Ranges} and {Chem Shift Ranges} tabs would allow you to make only restraints for specific assigned regions of your molecule or for specific shift ranges.

To calculate constraints for assigned peaks from the selected peak list simply press [Make Assigned Restraints]. After a short pause you will see the Restraints and Violations popup appear. This shows that you have one restraint set (a way of grouping related restraints and violations) containing one list of over 500 constraints. Click on the row of the constraint list in the central table and then click [View Restraints]. Note that you can also get to this point via M:Structure:Restraints and Violations.

The constraints popup will appear and in its table you will see the constraints listed, mostly as green coloured rows. Note some restraints also have following grey rows. These grey rows indicate restraints that are ambiguous, i.e. a possible connection between two different pairs of 1H resonances. Note that such ambiguous restraints can represent logical uncertainty (before an NOE is resolved) or real physical ambiguity where a peak is caused by two or more overlapping pairs of resonances.

Making Restraints From Unassigned Peaks

There is a second common way to generate distance constraints, which is to match the chemical shifts of resonances to NOE peak positions, thus generating potentially highly ambiguous distance constraints. Such constraints would typically be filtered to select only the correct contributing resonance pairs, by iterative structure generation and violation analysis in a program such as ARIA or CYANA. Accordingly the {Shift Match Tolerances} and {Network Anchoring} tabs in the M:Structure:Make Dist. Restraints popup allow you to generate such distance restraints for peaks which do not have assignments. To generate distance constraints by shift matching, click [Make Shift Match Restraints] This command uses the current settings, but {Chem Shift Ranges} and {Shift Match Tolerances} are only relevant for this command.

In the case of the shift-matching method potentially ambiguous distance restraints are generated by simply matching peak positions to close chemical shifts. In the case of network anchoring method chemical shift are also matched to peaks, but the ambiguous possibilities are refined by selecting only NOE assignments from amongst the possibilities that are supported by other, assigned NOEs or covalent structure. For example, given a peak that is possibly caused by several resonance pairs, including resonances A & B. Then if two resonances A & B are both known to be close to (or bound to) the same intermediary resonance, C then resonances A & B are likely to also be close, and the likelihood that they gave rise to the peak is increased.

Restraint Sets
Note that in general you are using the same restraint set as was used previously. You always have the option to put new restraints in a new restraint set, and indeed you should always use a new restraint set if any of the atomic assignments have changed. Each restraint set uses a frozen copy of the resonance-to-atom assignments at the time its restraints were created, so that you always have a record of what was really restrained, even if the resonances subsequently are reassigned to different atoms. By the same token, if you did not use a new restraint set after atomic assignments were changed the restraints would refer to the atoms from the old assignments not the new ones.

In the Restraints & Violations popup, which is hopefully still open (M:Structure:Restraints & Violations otherwise), choose the {Restraint Lists} tab and select restraint list 2 in the table and click [View Restraints]. You will now see that in contrast to the first list these restraints are highly ambiguous, with several possible resonance pairs for each restraint. If you click on the row of the first constraint "1:0" and then [Assign Peak],  the Edit Assignment popup will show the assignment possibilities and status for the peak that gave rise to this constraint. This illustrates that in the second dimension of this NOESY peak there are several resonances with shifts that closely match the peak position for that dimension. Note that the peak itself is not assigned to any of these, but that the correct contributing resonances may be assigned after a structural model had been calculated. Referring back to the Constraints popup, if you now click [View peaks] you will see that the NOE peak appears in a table from where you can then select the peak's row and click [Find Peak] to locate it in an appropriate window. Note that  a constraint may be linked to more than one peak, for example where there are symmetry related peaks that correspond to the same close resonance pair.

Next we will load a protein structure, corresponding to the spectra that are present in this tutorial. This structure has been generated using the program ARIA, but in general such a structure could be from any source including early structure calculation runs or an homology model. To load the structure select M:Structure:Structure Ensembles, set the molecular system as "MS1" then click on [Import] and select the file "G1.pdb" and finally [OK].  (If you don't already have the molecular sequence loaded at this point then say 'Yes' when asked if Phe and Tyr are to be considered fast-flipping on the NMR time scale.)  To see the structure select it in the table and click [View Structure].

Referring back to the restraints popup (M:Structure:Restraints & Violations {Restraints}),set the structure pulldown menu to "1", the structure we just loaded. You are now able to select any restraint rows (using <Shift>/<Ctrl> keys) and then click [View Selected On Structure] to illustrate graphically where on the loaded structure the restraints apply.

The controls for the structural viewer are as follows:

• Rotate with middle-click & drag.

• Zoom with the mouse wheel, or middle-click, <Shift> & drag.

• Move with middle-click, <Ctrl> & drag.

The mouse right-click is reserved for a menu and the left-click will be used for selection (in the future).

The same sort of functionality is present in the Edit Assignment popup (M:Assignment:Assignment Panel). - If you go back to the NOESY spectrum in Window 4, and assign a peak (by pressing <a> with the cursor over a peak), you can see assignment possibilities via the [Show On Structure] button. Also note that because we now have a structure the Edit Assignment popup will show distances between one peak assigned 1H resonance and the 1H possibilities in another peak dimension.

Refining Restraints With Structures

To make the whole process of comparing NOE peaks to a structural model efficient, Analysis has the M:Assignment:Link NOEs option. Select this from the menu and in the {Options} tab of the resulting popup set the NOE Peak List pulldown to N-NOESY, set the Mark Peak toggle on and finally double click in the "Use?" column of the lower table over the "No" for window 4, so that it changes to "Yes". Then go to the {Peak Assignments} tab. By clicking on the rows in the NOE peaks table you will see that several things happen automatically: The view of the selected windows (in this case the selected Window 4) zooms and marks the selected peak; the graphical structure view highlights the possible atom connections that the peak could represent; and the lower table of the Link NOE Resonances popup shows the structurally possible, shift-matched resonance pairs ordered in terms of shift and geometric distance. In the lower table clicking on an assignment row and [Assign Selected] sets that assignment for that peak. Also, selecting [Predict Peaks] will use the entered structure and the known chemical shift values to predict the positions of peaks, near to the selected real peak, which correspond to close resonance pairs. These artificial peaks are labelled with the structural distance and are contained in an entirely separate list to the real peaks (so nothing is contaminated and they can be removed easily).

Calculating Structural Violations

Lastly, given the structure that we have loaded into Analysis and the distance constraints that we generated from the NOE peak list, we can investigate the compatibility between the structural model and the constraint distances to find violations.

Return to the Restraints popup (M:Structure:Restraints & Violations {Restraints}) and select restraint list 1 from the pulldown menu and click [Calculate Violations]. After a short pause you will see that several of the constraint rows turn red. These indicate that the restraint distances are inconsistent with (i.e. violate) the loaded structural model. The peak assignments from which this constraint list was generated obviously needs some work, so that hopefully a new and better structural model can be generated. Many of the violations are small, but some are large, taking for example constraint " 20LeuHba - 26SerH", the mean violation is 1.44 Angstrom. If you select this row and click Assign Peak you will see that an assignment in the F2 dimension, to 76LysHba, is probably missing: the chemical shift "delta" is small and the distance given the structure is close. Clicking on the 76LysHba row in the Edit Assignment popup will ambiguously assign the peak. Hopefully when the next set of constraints is generated (and maybe a new structure calculated) the peak will represent a shared contribution from two close 1H pairs, and the 26SerH - 20LeuHba restraint will no longer appear too short. Returning to the restraint table, with the extra resonance assigned to the peak and same restraint selected click [Update Assignment From Peak], which will alter the distance restraint to reflect this new assignment.

At the menu option M:Assignment:Quality Reports you can access a suite of analyses which generate automated reports about the quality and completeness of your assignments. The first tab simply lists how many of each kind of residue and atom type you have resonance assignments for. The second tab shows statistics for how many NOE (and other through-space) connections you have across your molecular system.

While the first two tabs are dedicated to analysing completeness, the last two tabs are dedicated to finding mistakes and inconsistencies in assignments. These tables look at the same kind of issues, but one presents the information from a resonance point of view, and the other from a peak point of view. Most notably the assignments are checked for the following criteria:

• An assignment to a given atom is duplicated.
• The standard deviation for a chemical shift is large.
• If a peak position is far from the chemical shift average (irrespective of a shift's deviation).
• A resonance has an impossible number of covalently bound partners: e.g. an amide proton assigned to two HSQC peaks cannot be linked to both nitrogen resonances.
• The chemical shift for an atom assignment is unlikely given the known chemical shift distributions (BMRB).
• If peaks have unusual intensities or sign.

When something is potentially wrong, then cells of the tables are coloured red. A highlighted cell doesn't mean that something is definitely wrong, just that it warrants an explanation or further investigation. To find the peaks responsible for any aberration, simply click on the row in the table and click [Show Peaks]. In the resulting table you can assign or find any of the peaks in the various spectrum windows,