Analysis info tab

This contains the analysis parameters used to detect nuclei. If multiple datasets are selected, the background to each row will be green if all datasets have the same value, otherwise, the row will be white.

Images tab

Shows the images in which nuclei have been detected. The detected nuclei are highlighted with a yellow outline.

If image files have been moved, you can relocate them by double clicking the image folder and choosing the updated folder location.

Cells tab

This displays information on individual cells within a dataset. On the left is a list of all cells within the population.

The image channels with detected features (nucleus, FISH signals) can be switched between using the box on the lower left, under the cell list.

When a cell is selected, the source image will be displayed, with annotations for the centre of mass, segments, landmarks, and any FISH signals.

The orientation of the nuclei can be toggled using the Orient checkbox between actual (the same orientation as in the original image), and oriented (aligned according to the dataset rulesets).

Here is the outline displaying the image with the nucleus:

The centre of mass is a pink diamond. The landmarks are grey diamonds on the nuclear outline. Hover the mouse over them to see their names.

Segment boundaries and landmark positions can be updated. When you hover the mouse over the outline, a blue dot will be displayed. Click at any point to set it as a landmark or segment start location:

This is a dataset with a FISH signal, showing the nucleus and signal channels. The outline and centre of mass of the FISH signal is shown in red:

Nuclear charts tab

Displays measurements of nuclei within and between selected datasets.

Measurements tab

Mean, median, standard error of the mean, coefficient of variation and 95% confidence intervals for nuclear measurements.

Violin plots tab

This tab shows information on nuclear measurements for selected datasets in a series of violin and boxplots. The colour of each series in the boxplot reflects the dataset colour in the populations panel. To change colours, double click the dataset colour in the populations panel.

Wilcoxon tab

This tab shows the results of pairwise Wilcoxon rank sum tests (a.k.a. Mann-Whitney U-tests) of population medians for each of the nuclear parameters, with Bonferroni correction for multiple testing. P-values significant at the 5% and 1% are highlighted in yellow and green respectively.

The U statistic is given above the diagonal, and the p-value below the diagonal.

Magnitude tab

This shows the pairwise magnitude differences between datasets, to give a feel for the biological significance of any differences detected between samples. The value in each cell of the table for each measurement is the median value of the measurement in the dataset of the row, divided by the median value of the measurement in the dataset of the column.

Scatter tab

Allows comparisons between two parameters, via the drop down lists. The Spearman's rank correlation coefficient for each selected dataset is given in the table to the left. The nuclei can be filtered using the 'Filter visible' button.

Nuclear profiles tab

The profiles tab displays the morphological profiles calculated for datasets.

At the top of each chart are display options:

• Normalised: when ticked, all profiles are stretched to the same length. When unticked, nucleus profiles are shown at their actual length, corresponding to the perimeter length of the nucleus (the median profile length is set to the median of the lengths of the nuclei in the dataset)
• Left: if the displayed dataset(s) are not normalised, they are aligned to the left edge of the chart
• Right: if the displayed dataset(s) are not normalised, they are aligned to the right edge of the chart
• Show annotations: if selected, show segment and landmark names (only applies when a single dataset is selected)
• Show nuclear profiles: if selected, show individual nucleus profiles (only applies when a single dataset is selected)

If a single dataset is selected in the populations panel, the profile tab displays the profiles for all nuclei in the dataset in light grey, the segmented median profile in the chosen palette colours, and the interquartile range about the median in dark grey.

If multiple datasets are selected in the populations panel, the profile tab displays the median profile in the chosen dataset colour, and the interquartile range about the median in alighter shade of the dataset colour.

It contains the following tabs:

Angle profile

Calculated by measuring the angle between the three border points from the point window size pixels behind to this point, to the point window size pixels ahead of this point.

Diameter profile

The distance from each border point, through the centre of mass of the nucleus, to the opposite border. Expressed as a fraction of the maximum diameter per nucleus.

Radius profile

The distance of each border point from the centre of mass. Expressed as a fraction of the maximum radius per nucleus.

Variability

The variability plot shows the size of the interquartile range along the median profile, and therefore shows the regions of the profile that are most variable within the dataset. The drop-down list at the top allows the variabilty of each profile type to be selected.

Nuclear signals tab

This tab displays information relating to FISH signals.

Overview

This shows a table with average information about each signal group that has been detected, and a chart with the outline of the consensus nucleus (if refolded) and approximate positions of each signal.

The signals from each nucleus are drawn as a dot at the centre of mass of each signal. Tick the 'Show signal radii' option to draw a translucent circle on each signal with an equal area to the measured signal. Hover the mouse over a signal to show the nucleus image from which it was detected.

Note - a more accurate mapping of signals onto the consensus can be achieved using signal warping.

Signal groups within a dataset can be merged together using the 'Merge signal groups' button. Merged signal groups do not affect the original signal groups - they duplicate the original signals and add them to a new group.

Detection settings

This shows the settings that were used to detect each signal group in the dataset. Double click the signal group colour to change the colour.

Signal counts

Displays the number of signals detected per nucleus. Datasets can be filtered to select nuclei with a given number of signals using the 'Filter nuclei' button.

Violin plots

This shows violin and boxplots for each of the signal measurements.

Scatter

This allows scatter plot comparisons between signal measurements, with Spearman's rank correlation coefficients shown at the left. As with all scatter panels, the datasets can be filtered on visible values using the 'Filter visible' button.

Shells

Displays the resuls of shell analysis. Shell analysis can be run via 'Dataset' > 'Add' > 'Shell analysis of signals'.

If the selected datasets contain a single signal group, the chart will display a bar chart of the percentge signal detected in each shell. If multiple signal groups with shell results are present, the chart will instead show a heat map, in which the intensity of each square corresponds to the percentage signal in each shell. You can think of it as looking at the bar charts from above.

The upper table shows whether the shell results are significantly different to a random distribution via a chi square test.

The lower table shows pairwise comparisons between any shell results in the selected datasets, again by chi square. The pairwise tests require the shell results to have the same number of shells.

Colocalisation

Only active if there is more than one signal group present in a dataset. This shows the distances between closest signal pairs. For each signal group within a dataset, signal pairs are calculated using the following method:

• Signals within a nucleus are identified
• The distances between all signals of different groups is calculated as the distance between the centres of mass.
• The pair with the shortest distance is recorded and removed.
• The process repeats with the remaining signals
• Any remaining unpaired signals are ignored and the paired distances are reported.

Warping

Displays the results of signal warping. The table on the left shows all warped images in the selected datasets, and their options. Click a warped signal to display it in the chart on the right. Select multiple warped signals to create an overlay.

The 'Pseudocolour signals' checkbox allows you to toggle between greyscale signals and the pseudocolour shown in the table. Double click the colour box to change the pseudocolour. Note that this is distinct from the signal group colours in the 'Nuclear signals tab' - you can have the same signal group warped with different options in different colours.

The slider allows thresholding of the warped signal.

The 'Export image' button allows you to save the warped image to a file.

The 'Full MS-SSIM*' button opens a comparator showing the full structural similarity scores between any compatible warped signals in any datasets currently open.

Nuclear segments tab

The segments tab displays information on the segments within the median profile of the selected dataset(s).

Violin plots tab

The violin plots tab shows the range of segment lengths in the desired scale for all selected populations. If multiple datasets are selected, they must have the same number of segments for a meaningful comparison to be made.

Wilcoxon tab

The Wilcoxon tab shows the Wilcoxon rank-sum tests for pairwise comparisons between datasets for each segment.

Magnitude tab

The Magnitude tab shows the the pairwise magnitude differences between datasets for each segment.

Comparisons tab

The Venn tab compares overlaps in nuclei between datasets, presented as a table.It is primarily designed to allow comparison of different child datasets after filtering or clustering.

Detailed Venn presents the Venn data in a pairwise format, with more detailed breakdown of the numbers of shared and unshared nuclei.

Clusters tab

The clusters panel shows information on clustering analyses that have been run previously. When cluster groups are available, the analysis parameters for the groups are shown in the table.

Merges tab

If a selected dataset is a merge of other datasets, this panel displays the analysis parameters of the source datasets. As with the main analysis info tab, matching values are highlighted in green for each row. It also allows you to extract the source datasets back out of the merge. See also merge instructions.

Editing tab

This allows segments and landmarks to be updated within root datasets. The consensus nucleus for the selected dataset is displayed, with annotations for the landmarks. The landmarks are grey diamonds on the nuclear outline. Hover the mouse over them to see their names.

Segment boundaries and landmark positions can be updated. When you hover the mouse over the outline, a blue dot will be displayed. Click at any point to set it as a landmark or segment start location. All cells will be updated to their best fit to the new location:

If you want to merge or split segments, use the buttons at the top of the panel. Merged segments can be unmerged; the unmerge button will be greyed out unless there are merged segments available.

Angle window size explorer

If you want to see the effect of changing angle window sizes, click Explore window size. This will open a new window that allows the effects of altering the angle profile window size to be simulated across a range of values. No changes will be made to your actual dataset.

Enter the smallest angle window size to test (0-1), the largest to test (0-1), and the step size between them, then click run. The angle profiles generated for each of the window sizes will be displayed on the chart. The profile colours form a gradient from blue (lowest window size) to red (highest window size):

If you want to change the window size in your actual dataset, click the Set window size button.