How sensitive is Pyxa? How many genes in the panel can be reliably detected?

For the 235-plex Mouse Neuron Cell Typing panel, we assess sensitivity by asking how many panel genes are detected above background.

In 3D Pyxa data, 227 of 235 genes (96.6%) were detected above a 99.5% confidence noise floor. We set the noise floor with ERCC negative control probes run alongside the panel. The 99.5% confidence limit was calculated from the ERCC counts using a Poisson model, and a gene is called detected only if its counts (normalized per 50k cells) exceed it.

The detected genes span roughly 2.4 to 5.6 log10 counts per 50k cells, so the panel captures both highly and lowly expressed transcripts. The few genes that fall below the threshold sit in the low-abundance tail of the distribution.

Please note this metric describes detection above background across the panel. It is not a per-transcript capture efficiency. Performance can also vary by tissue, sample quality, and panel design. If you share your tissue type and target genes, we’re happy to discuss what to expect for your experiment.