Graphs
A: ioGAS can only display up to 2000 data points per group as specified in the Line Plot tool. If there is a group with greater than this number of data points then the data will not be displayed in the line plot. In older versions than ioGAS 4.4 the limit is 1500.
A: Make sure you haven’t turned off gridlines for the plot using the show/hide gridlines icon (right hand tool bar, usually at the bottom of the bar).
A: ioGAS displays the sample number and the values for the displayed axes for many of the plots in ioGAS. The cursor must be over the data point of interest for these values to display. Only one value is displayed at a time. For example, on the map view, the sample ID, northing, easting, and variable value are all displayed in a pop-up text box when the cursor is hovered over a plotted point. If this information is not displaying, check that there are columns set as ‘Sample ID’, ‘North’, ‘East’ in the Column Properties dialog. If you wish to view other columns as metadata, go to the Data>Select Labels menu option and add some more text columns to display.
A: Make sure you have variables selected in the Data>Select Variables dialog first. Different plots require different numbers of variables to be selected, eg. two for a scatterplot, three for a tri-plot, etc.
A: If you are making groups of data invisible or visible via the Attribute Manager, the scaling of the graph will change to reflect the visible data range. As this may not be desirable, the padlock (lock axes) icon on the right hand tool bar can be used to prevent automatic scaling.
A: You have to select an attribute in the Attribute Manager first before selecting any point or polygon to apply. Add a new colour, shape or size combination and hightlight each one so that it appears in the Attribute Manager title bar. When an attribute manager combination is ready, the Select Point or Select Polygon icon should work.
A: Make sure that the column type for each variable matches the plot type being created e.g. text column types will not plot in box plots.
A: Make sure that all attributes in the 3 tabs (shape, colour, size) are set to visible in the attribute manager. Also, check that the samples have a valid data entry for all the variables in the plot axes.
A: It is not possible to display a log axis on a plot if the data being plotted contains 0, negative or text values. When you click on the Log icon and nothing happens check the bottom left hand corner of the status bar and if this is the problem you should see a red message in the bottom left hand corner of the screen which says “Unable of Log zero, negative or text data”.
To replace 0 or negative values in your data use the Validate>Data Doctor utility.
If you hover the stylus over the screen a black point follows the location of the tip and even when this passes over a point the label is displayed but not necessarily all the time. If you actually place the stylus right on the point and leave it there for a moment it shows the label BUT you have to be right on the point and leave the stylus there for a moment for ioGAS to register the location.
A: Right-mouse click on the individual plot and select Properties. In here you can change the plot or axes labels, axis ranges, plot background colour and more. Note that it is not currently possible to save these changes to the ioGAS file.
A: Yes. Make one window the active window and then go to Windows>Remember Window Size. This will hold the current window size in memory so if you then select another window and choose Windows>Restore Window Size this will re-size the active window to the stored dimensions.
A: A transformation that is trying to estimate a value for lambda to de-skew data into a more normal distribution. See Power Transformation for more information.
A: The first variable in the list is used as the X axis and all of the other variables are then plotted against it in the XY scatterplot graphs. This behaviour does not apply to scatterplot matrix diagrams or map views.
A: If you created the plot with the XY scatterplot icon, use the YX scatterplot icon. The capital represents the axis on which the first variable in the variable select dialog will be placed.
A. Use the Copy Entire Window at High Resolution window button at the bottom of the Right-Hand toolbar. This will copy the entire window as a bitmap image to the Clip Board which you can then paste into an image processing program (eg. Paint or Irfanview, Paint Shop Pro, etc) and then print to your plotter from here.
A: Yes. Before creating the scatterplot matrix go the Window menu and select Scrolling Window. Make the scatterplot as usual and there will now be up/down and left/right scrollbars available to scroll around the plot and the individual plots should be displayed at a higher resolution.
A: The only way to modify these very long labels at present is to either click on an individual plot and choose Properties and modify the name in here or to do it for each column in the Column Properties dialog prior to creating the scatterplot matrix.
A: ioGAS has a number of summary tools available. Both parametric and ranked based statistics can be plotted per colour attribute group. The samples points can be hidden to make viewing these summary statistics easier.
A: Make sure the column is correctly aliased in the Column Properties dialog. Spider transformations rely on aliasing, in order to correctly identify elements and their units.
A: Use the Copy button on the Right Hand Side toolbar when the Spider Plot window is opened to copy to clipboard and then paste into WordPad, NotePad, etc.
A; Yes. See Create Spider Plot for more information.
A: To switch from displaying solid mean data to all data in a spiderplot use the second icon down on the RHS toolbar called “Change visibility of data points”. If your dataset has a large number of rows ioGAS automatically displays the spiderplot using the data means as it difficult to view the individual samples when there are a lot of them. If you don’t want to display the mean data lines on top of the individual data rows then click on the “Change Parametric Statistical Summary” icon a couple of times to remove them from view (3rd icon on RHS toolbar). To summarise: the 2nd icon down turns individual data points on and off, the 3rd icon down cycles through showing Solid Means, Solid Mean +/- 1 sigma or no mean lines. The 4th icon down showing Percentile summaries works in the same way. Sometimes you need to play around with each icon to see exactly what it does and click on it a few times to see the changes and how they work together.
A: The data groups which are used to display the mean lines are taken from the colour attributes only. It is only when the individual data rows are displayed on the Spiderplot that you can see the colour/shape/size attribute for each data point.
A: The pie chart plots the colour attributes that are set in the Attribute Manager. Create colour attributes using manually selected groups or automatically assigned from text or numerical variables.
A: N-Score and probability are the same. The n scores are the z scores from a standard normal distribution ie. mean 0 and sd 1 (hence n score, not z score). Data is plotted against the n scores from that distribution. If you picture the bell curve, these are the numbers on the x axis in units of sd. If you replaced these by the area under the standard normal curve they would be probabilities, so, an n score of 0 (middle) would have an area to the left of 0.5 so a probability of 0.5. An n score of 3 would have a probability of 0.99 or so. To view Probability values on the X axis click on the Change X ticks button on the RHS toolbar.
A: You need to attribute the different populations first and then view them using the Statistics tool. Open the Attribute Manager and add a colour attribute. Select this colour attribute and then go back to the Probability plot window. Use the Select Attribute Polygon from RHS toolbar to draw a polygon around a population. Repeat this process for any other populations and then select Analysis>Display Summary Statistics. Scroll down to see statistics by each colour population group.
A: A split probability plot subdivides the data according to the colour attribute groups, and plots a normal probability plot for each group for the selected variables, on the same diagram (with the respective colour from the attribute dialog). This allows quick comparison of distributions of subtypes in the data e.g. soil versus stream sediment, or drill hole 1 data versus drill hole 2 data. If a group has a normally distributed sample population, then they will form a straight line.
A: The histogram chart uses the current colour attributes to subdivide the data. This allows quick visualisation of the relative contributions by different subgroups to the overall sample count. The stacked mode merely displays the coloured bars on top of each other, and thus the totals for each bin would be the same as for a histogram without colour groups (ie only ‘default’ set in the colour attribute dialog). The layered view plots each group 1 behind the other, with the colours made semi-transparent, and is generally only useful for 2 or 3 groups. The number of bins can be modified using the up and down arrow icons, while individual colour bins or the entire bar can be selected for attributing using the top two icons.
A: Parallel coordinates plot is a technique for representing high-dimensional data.
To show a set of points in an n-dimensional space, onto a 2D surface (ie. the computer screen), the graph is drawn consisting of n vertical, equally spaced parallel line segments. The lines themselves are not visible in ioGAS, but have the sample points plotted as dots onto them, according to that sample’s value in that variable. The number of vertical lines depends on the number of variables selected in the Select Variables Dialog. Then the sample’s points on the various vertical lines are linked with a polyline.
Part of the value of parallel coordinates is that certain geometrical properties in high dimensions translate into easily seen 2D properties. For example, a set of points that lie on a line in n-space will translate to a set of polylines in parallel coordinates that all intersect at a common point; two sets of points which occupy distinct spatial clusters will form ‘knots’ of lines that don’t intersect.
Selecting more variables and creating another PC plot is equivalent to adding more dimensions to the parallel coordinate plot.
A: ioGAS has a number of summary tools available. Both parametric and ranked based statistics can be plotted per colour attribute group. The sample points can be hidden to make viewing these summary statistics easier. To see the various tools available and how they work click here.
A. Hover cursor over main box and view in tooltip. Copy all Box Plot statistics to text editor such as Word or NotePad using the Copy button on the RHS toolbar.
A: Yes, use the Point Attributing On/Off button on the RHS toolbar.
A: If you are using Drillholes for example to display on your stacked line plots the order in which they appear in the line plot tool is the same order they are appear in the original data file. If you want to change the order then you will need to modify the original data file and re-import it into ioGAS.
A: In the Column Properties dialog make sure that the numeric variables in your dataset are aliased and have a green dot next to them. Click on the Select Mineral Groups icon on the RHS (Right-Hand Side) Toolbar in an open plot window and make sure there is at least one mineral composition checked to display. Use the Show/Hide Mineral Compositions icon to display the minerals in the plot window.
A: The chemical composition of each of these minerals is as follows:
Chlorite 1 = Mg5Al2Si3O10(OH)8
Chlorite 2 = Mg6AlSi3O10(OH)8
Chlorite 3 = Fe2.5Mg2.5Al2Si3O10(OH)8
In ioGAS version 4.5, the composition of each of these minerals is displayed when you hover your mouse over the mineral name in the selection window.
Alternatively, to find out what the chemical composition is for these and other minerals displayed in a plot window click on the Show/Hide Mineral Compositions button on the RHS toolbar and it will cycle through displaying abbreviated mineral name, full mineral name and chemical formula for each node.
A: It is possible to create a density plot for the attribute groups separately if you turn the visibility of all the other groups off in the Attribute Manager.
A: The recommended way of dealing with zeros in ioGAS is to use the Validate>Data Doctor tool which has a range of options for how to treat zero, negative and other ‘special’ values. Then all the plots and statistics are ‘clean’.
A: Plots with all values >0 can be logged by using the log button on the right hand side toolbar (including data density plots).
A. Copy all Box Plot statistics to text editor such as Word or NotePad using Copy button on RHS toolbar.
A. Use the Copy Entire Window at High Resolution window button at the bottom of the RHS toolbar when the scatterplot matrix window is open. This will copy the entire window as a bitmap image to the Clip Board which you can then paste into an image processing program (eg. Paint or Irfanview, Paint Shop Pro, etc) and then print to your plotter from here.
A: It sounds like the data limits for the installed ioGAS executable file have been reached. Although ioGAS could successfully create the SPLOM copying to the clipboard was unable to be completed. For more memory intensive operations such as creating large SPLOM plots (and in this case copying to clipboard), gridding or even just importing large files it is possible to load up additional ioGAS executable files which enable more machine memory to be utilised. See Data Limits for more information on how to do this.
It is not possible to embed the legend into a plot window but you can see the legend in a separate window by clicking on the blue circle icon on the main toolbar.
You can either just use the zoom tools on the RHS toolbar and zoom in on your cluster. The 4th, 5th and 6th icons can all be used to achieve this.
If you want to specify a range to display you can right-mouse click on the plot and choose Properties. Under the Plot tab you can set a minimum and maximum axis range here.
It is not possible to set a default font at this stage. It is possible to change individual plot font, size, etc using the Graph Properties option from the right-mouse click pop-up menu for individual plots.
If you hover over the line in the plot it gives you the equation in a tooltip. You can’t automatically add it to the plot. If you use the regression tool (from the Analysis menu) the reports contains the regression equation.
A: In ioGAS version 4.5, outlier points in Box and Whisker plot can be selected by clicking on the Attribute Polygon button (same button as in scatter plots), then drag a box around the points of interest to extract outlier points.
In version 4.4 and older, it is not possible to select the outlier points directly on the B&W plot, however if you go to ‘analysis>Auto-Attribute Tukey Outliers’, it will run the tukey algorithm and then automatically attribute any outlier data points.
If you click on the “Lock” icon on the plot window RHS toolbar (3rd one from the bottom) before you make a row invisible it will lock the view and the window will not re-size when you turn the visibility of data on/off. You can then use the “Unlock” icon to go back to the way the plot window normally works which is to zoom to the extents of the visible data.
The algorithm was not changed but the outlier values that we report in the tooltip has changed. In v4.3 we reported the near and far outlier min and max SAMPLE values and in v4.4 we have changed this to report the near and far outlier THRESHOLD values.
Aside from using your drillhole data with any of the standard ioGAS plots and univariate/multivariate analysis tools there are two main tools that are designed specifically to be used with drillhole data. These are the 3D spatial plots and the Line Plot tool. The 3D spatial plot works just like the Attribute Map but for drillhole data which already has XYZ values for each sample interval in the dataset. You can apply colour/shape/size attributes to the downhole sample intervals and even export the attributed data in 3D DXF format to use in other mining programs.
The other tool which you may find useful for working with drillhole data as well is the Line Plot tool. This enables you to display each of your drillholes in a continuous line plot which can be stacked for easy comparison between holes. It is possible to attribute the line plot data points with other variables as well.
A. Click on the Line Plot icon on the main toolbar. Select the Hole_ID column in your dataset as the grouping column and a depth or Z coordinate column as the continuous variable. Leave the variable format as numeric.
In the next dialog select the drillholes you would like to display and how you would like to display them, eg. stacked on top of each other, tiled, etc.
For a full description of how to use the Line Plot tool and a list of the various Line Plot options click here.
You can make box-plot graphs comparing soil and sediments for the same parameter using the following steps:
1. Select the variables you want to look at it the Data>Select Variables window
2. Use the Colour attribute in the Attribute Manager to colour the samples by sample type – soil and sediment (they will need to be in the same file)
3. Select Graph>Tukey Box Plots (common Y axis) menu option
A: Unfortunately, at the moment it is not possible to remove the word [Locked] from the axis label when you have the axis locked. Thus the only way to remove the word [Locked] from the axis label of XY graphs is to unlock the axes. Note that this will distort the shape of line and polygon features in XY diagrams.
A: There is an issue with importing vector files into an image editing program in which shapes don’t retain their original form as seen in ioGAS. This was quite apparent when circular shapes are being imported as diamond or square shapes. This only happens when you copy graphs straight from the right-click menu in ioGAS (i.e. Right click>Copy>Copy EMF or Copy EPS) and into an image editing program. By saving the plot as an EPS or EMF format from the right click menu (Right click>Save>Save EPS or Save EMF) and bringing that in to an image editing program, the shapes will now look correct.
A: A spiderplot works by dividing the analytical data with the normalised value specified in the plot’s xml file. A spiderplot xml file consists of a list of elements and the normalisation value for each element. For example, if you want to create a spiderplot that normalises to the following values:
La of 50 ppm, Ce of 100 ppm, Dy of 5 ppm
You will then need to edit the values bracketed by <m> </m> (shown in bold in the text below). The spiderplot will divide by these values to normalise the plot.
<transformation_mx_c name=”test”>
<Comment>test</Comment>
<Reference>test</Reference>
<element name=”La” unit=”ppm”>
<m>50</m>
<c>0</c>
</element>
<element name=”Ce” unit=”ppm”>
<m>100</m>
<c>0</c>
</element>
<element name=”Dy” unit=”ppm”>
<m>5</m>
<c>0</c>
</element>
</transformation_mx_c>
For more details on how to set up custom spider plot, please refer to the following guide: http://iogas.net/wp-content/uploads/2012/03/Create-Spider-Plot-Normalisation.pdf
A: This is caused by having too many variables selected for the scatterplot matrix, thus causing ioGAS to try to fit them into a single window. A solution is to reduce the number of variables to display. This would increase the individual plot size.