Using the GUI: A tutorial

Starting in v0.3.7, you can launch a GUI tool that provides some basic capabilities for a small number of sequences (we plan to extensively expand this tool in future versions).

To run the GUI, you must have two additional dependencies, pyside6 and qt_material installed. Next, from the command line, type::

AntPack-GUI

This is in contrast to::

AntPack-CLI

which runs the command line interface. You should see the following screen:

GUI display

Note that at this point, only the “Sequence Review” is implemented (more to come later). You can select a numbering scheme using the file menu, then you can either view a single sequence together with the closest corresponding VJ / germline genes from humans or mice, or you can add and view as many sequences as you want and see how they compare.

For example, we can add a sequence by clicking on add sequence then pasting it into the appropriate blank and selecting appropriate options:

Add sequence

for more about some of these options, see the AntPack numbering tutorials. You can view a mixture of single and paired chain sequences as illustrated below – the sequences with only one chain will show up as heavy or light only under the appropriate view. Notice that CDRs for your currently selected numbering scheme are marked using “_”.

Add sequence 1 Add sequence 2

You can also compare a single paired or single chain sequence with the closest VJ / germline sequences from human or mouse as illustrated. This feature is currently supported only if the IMGT numbering scheme is selected and uses assignment by percent identity only at present. If multiple V/J genes have the same percent identity to the input, all will be shown.

VJ comp

Notice that the GUI (at present) is not so powerful or flexible as the Python API and offers fewer options; currently, it’s best suited for quickly comparing a handful of sequences. If you want to do a standard analysis on a much larger group of sequences, you may prefer the command line interface, and if you want a more powerful and flexible tool to design your own workflows the Python API is likely preferable.