Inquiry Regarding Differences in Gate Values Between SCIMAP Versions. #129
Description
Dear SCIMAP Team members,
I hope this email finds you well.
I am currently using SCIMAP for my data analysis, and I have encountered an issue with the calculation of gate values across different versions of the software. Specifically, I have observed that the gate values differ significantly when using SCIMAP 2.1.3 and SCIMAP 2.3.1, even when using the same input data and running identical code.
Issue Description:
In both versions, I applied the following code to preprocess and rescale the data:
import scimap as sm
adata = sm.pp.rescale(
adata,
gate=None,
log=False,
imageid='Image',
failed_markers=None,
method='all',
verbose=True,
random_state=0
)
However, when I examined the resulting gate values for the same dataset, I found that the gate values produced by SCIMAP 2.1.3 and SCIMAP 2.2.11 were different. For example, the gates in SCIMAP 2.1.3 were around 0.2 to 1.0, while in SCIMAP 2.2.11, they ranged from 0.6 to 2.0. This discrepancy is concerning, as gate values are critical for phenotyping and downstream analysis.
Questions:
What changes have been made between SCIMAP 2.1.3 and SCIMAP 2.2.11 regarding the calculation of gate values?
Why do the gate values differ between these two versions, even when the same data is being used?
Which version of SCIMAP should I trust when it comes to the gate values for accurate phenotyping?
Are there any specific preprocessing steps or settings that could explain the differences in the results between these versions?
As gate values play a crucial role in determining the phenotypic profiles in my analysis, I would greatly appreciate your guidance on which version provides more reliable results, and whether I need to make any adjustments to ensure consistent outputs across versions.
Thank you very much for your time and assistance. I look forward to your response and any insights you can provide.