Open ke4 opened 4 hours ago
Here is a previous ticket that tried to provide a solution on the backend.
Make cell plots k options compatible with unsupervised cluster ids
We need to investigate it.
I can also reproduce it in our local env with this experiment:
http://localhost:8080/gxa/sc/experiments/E-GEOD-99058/results/marker-genes
In the Single Cell Expression Atlas the cluster labels are not in the correct sorting order in the marker gene heat map.
Steps to reproduce:
Here is the full URL: https://www-test.ebi.ac.uk/gxa/sc/experiments/E-MTAB-10448/results/marker-genes
Expected result: The cluster labels are sorted naturally.
Actual result: The cluster labels are not sorted naturally. Look at the screen shot.