Cluster4x

Cluster4x looks at multi-datasets and partitions data into datasets which behave in similar ways. It is interactive, allowing you to fully explore your data set. It can compare your datasets using amplitude differences or C-alpha atom positions.

Cluster4x is often used to increase the power of PanDDA to find hits in your dataset. Often pre-clustering with PanDDA will lead to more hits in your datasets. The published paper shows an increase in the hit rate by over 65% when re-analysing the published data for a PTP1B fragment screen.

Above: screenshot of GUI in action for a previously published dataset, BAZ2B, showing two distinct clusters which are best handled separately in PanDDA.

Downloading cluster4x

On most flavours of Linux, you can easily download cluster4x directly using snap.

How to use cluster4x

See this video for a theory and software tutorial video. Skip to 2:32 if you don't need to watch the theory section.

How to cite cluster4x?

Ginn, Helen M. "Pre-clustering data sets using cluster4x improves the signal-to-noise ratio of high-throughput crystallography drug-screening analysis." Acta Crystallographica Section D: Structural Biology 76.11 (2020).

Bibliography entry for LaTeX:

@article{ginn2020cluster4x,
title={Pre-clustering data sets using cluster4x improves the signal-to-noise ratio of high-throughput crystallography drug-screening analysis},
author={Ginn, Helen M},
journal={Acta Crystallographica Section D: Structural Biology},
volume={76},
number={11},
year={2020},
publisher={International Union of Crystallography}
}

Installation from source

The additional program cluster4x is installed with Vagabond, if the Qt/Qt5 libraries are successfully found during installation. If they are not found, both cluster4x and the Vagabond GUI will not be installed.