You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Subseting cells in any plot is a very nice addition, however it causes errors in the heatmap when selecting groups where none of the cells have the expression for a given gene. The scaling + cluster rows does not work and the output is : An error has occurred. Check your logs or contact the app author for clarification.
I guess the row where all is 0 generates some NA values that should be ignored for clustering the rows
If that takes some time to implement a simple warning indicating what is happening would already be nice (similar to the "Only 1 group present, unable to plot!" warning when not enough groups are selected) . As it took a bit to understand from where the error was comming from, but once you know it, it is as simple as untiking the "cluster cells" to still see the expression.
The text was updated successfully, but these errors were encountered:
NadineBestard
changed the title
Heatmap/bublle plot gives unspecific error when subsetted for groups that have 0 expression for a gene
Heatmap/bubbleplot gives unspecific error when subsetted for groups that have 0 expression for a gene
Jan 26, 2023
We recently identified a related issue where NA were populating on bubbleplots clustering and throwing an error. Like you we also identified that error was resolved by unselecting the cluster rows. We fixed by forcing NA to 0, perhaps this might resolve the issue you describe as well? You could try our fork at https://github.com/VCCRI/ShinyCell (which is also a pull request on this parent repository). On our fork there is also additional tab allowing bubble plots on one gene which we find quite useful.
Subseting cells in any plot is a very nice addition, however it causes errors in the heatmap when selecting groups where none of the cells have the expression for a given gene. The scaling + cluster rows does not work and the output is :
An error has occurred. Check your logs or contact the app author for clarification.
I guess the row where all is 0 generates some NA values that should be ignored for clustering the rows
If that takes some time to implement a simple warning indicating what is happening would already be nice (similar to the
"Only 1 group present, unable to plot!"
warning when not enough groups are selected) . As it took a bit to understand from where the error was comming from, but once you know it, it is as simple as untiking the "cluster cells" to still see the expression.The text was updated successfully, but these errors were encountered: