No Chromosome has been selected. First select a chromosome from the Connections
Chromosome conformation is in its relative infancy compared with other molecular
biology tools to explore cellular attributes, and the volume and complexity
of these data pose considerable challenges. Therefore, we developed Rondo,
a web-based interface for visualizing Hi-C data, which minimizes visual
noise by hierarchical clustering and grouping individual connections into
larger groups that are progressively separated as more detailed information
is provided. Rondo also displays gene tracks and epigenome data sets, such
as histone modification marks, allowing biological interpretation of Hi-C
Here you can select which experimental data to view from the dropdown.
You can also select a second experiment to compare against.
Information about the experiment is provided in the panel below.
To determine if an experimental connection is statistically significant, they must be clustered into
a given bin size. Then each cluster is compared to each other cluster to see if it is over-represented or under-represented
when compared to what is expected for that distance. A smaller significant bin size will give less results, but they will be more precise.
To give a trivial example, if there are connections from Chr1:100k to Chr2:1500k and from Chr1:800k to Chr2:1800k then at 100k bins
they look like single connections and may be filtered out as insignificant, however when clustered to 1M there are now multiple connections so they
may be considered significant.
The statistics panel shows the total number of connections for the current view broken down into 3 categories.
Internal connections: show the number of connections within a single cluster. They are not visible at this level of zoom,
but may appear when you click on that region and zoom in.
External connections: show the number of connections from the given view region to other clusters outside the view region.
Cross connections: show the number of connections between different clusters within the same view region.
This panel allows you to filter to only the regions you care about. You can add regions to the text panel in the format of
[#chromosome]:[#start]-[#end]. e.g. Chr1:12345-23456
Press the APPLY button and then these will then appear in a table where you can also delete filters if necessary.
If you want to add additional filters, press the + button. If you want to remove all filters, press the CLEAR button.
You can specify that connections need to hit ANY (the default) of the filter regions, or ALL of the filter regions.
You can also specify a Leeway of how much room on either side of the filter is valid. e.g. If the filter is set to Chr1:10000000-20000000
and there is a connection touching Chr1:20500000, normally that would not be included. However, if the Leeway was set
to 1M, then that would be included.
You can specify a Leeway of how much room on either side of the filter is valid. e.g. If the filter is set to Chr1:10000000-20000000
and there is a connection touching Chr1:20500000, normally that would not be included. However, if the Leeway was set to 1M, then that would be included.
There are two histograms in this panel:
Statistical Quality Histogram: shows the frequency of connections based on the q-value.
Connection Histogram: shows the frequency of the number of connections of each cluster.
You can filter the connections based on various criteria:
Q value min: filters out all connections that have a higher (worse) Q value than the required Q value specified on the slider. The Quality histogram is updated to reflect changes to the required minimum Q value.
Q value high confidence: Connections that have a higher (worse) Q value than the high confidence value set on the slider are coloured semi-transparent. Better quality connections that meet the high confidence slider are shown as solid colour.
minimum connections: If the number of experimental connections between two clusters are under the value specified in the slider, they are filtered out.
Ignore Trans: when enabled, it filters out all trans connections from the view.
Ignore Neighbours: filters out connections to neighbouring clusters. e.g. connections from Chr1:1M-2M to Chr:2M-3M will be filtered out if the cluster size is 1M.
This panel is enabled once you zoom down to the gene level (around 1 Mb view range). A table shows each protein coding gene
found. A dropdown at the top allows you to choose between how these genes are found.
Regions: find all genes within the given regions displayed.
Connections: find all genes overlapping the connections displayed.
Search Terms: find all genes that are entered as the search terms.
You can also change the number of items displayed on the table with the dropdown in the top left. The gene list can be exported
with the controls in the top right. An interaction diagram will display below showing how these genes interrelate. This
diagram is taken from the
String DB with a stringent required score of 950 (meaning show only strong connections).
There are various graphical settings that can be applied for user preference, or for exporting for presentations or publications.
This panel also provides the export menu. The graphical settings include:
Background Colour: default is white, however black can be striking for presentations.
Legend Colour: Default is grey, can be white or black.
Detail: can be either normal (default) or high. High will download and display the next higher resolution information
normally obtained only by hovering over a section or zooming into it. While this provides more detail, it is slower
Ruler Labels: can be either more (default) or less. More is fine for display and exploration, however less could be better
for presentations and publications.
There are numerous formats in which you can export images:
SVG: vector output that scales well for print. Can be imported into Adobe Illustrator.
PNG: a simple graphical file format.
PDF: a PDF export.
All the connection data can also be exported using the following options:
all (bedpe): A
bedpe file format which allows both ends of the connection on the same line.
all (bed): A
bed file format which requires two lines per connection.
unique (bed): A
bed file that does not repeat regions that are hit from multiple connections. This is useful to analyse connection
dense regions in external tools.
matrix (tsv): A matrix of connections grouped by the cluster size in a tab separated value file (TSV). Warning, this
can be a really big file.