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
Hello,
Thank you for the very helpful package! I am having an error when I try to run ambientContribMaximum and am unsure whether I have misunderstood its application or if I could have come across an error.
I have multiplexed data with single cells from n samples in a single raw matrix. I calculated the ambient contamination with ambientProfileEmpty(raw_matrix, good.turing = TRUE), generating a vector. I wanted to estimate the ambient contamination for all my multiplexed samples, and therefore tried running ambientContribMaximum(samples, ambient, mode="scale") where samples is a matrix of n samples. I assumed I could estimate the ambient contamination for multiple samples or clusters within a run given this line in the documentation of ambientContribMaximum: "Hence, the function works best when y contains aggregated counts for one or more groups of droplets with the same expected expression profile, e.g., clusters of related cells."
However I receive the following error when I try to run this:
Error: BiocParallel errors
element index: 1
first error: 'to' must be a finite number
In addition: Warning message:
In .maximum_ambience(y = y[, i], ambient = A, ...) :
'y' and 'ambient' do not have the same feature names
I can run it if I ensure ambient has the same number of columns as y (by copying ambient multiple times in a matrix), but I am wary about whether this is the most appropriate course of action. Have I misunderstood the application of ambientContribMaximum/is there anything I should do differently? Do you have any suggestions as to how to accomplish my goal?
Thank you in advance!
The text was updated successfully, but these errors were encountered:
Can you provide a minimum reproducible example? For example, example(ambientContribMaximum) works fine for me. What is different about your data that the same code doesn't work for you?
Hello,
Thank you for the very helpful package! I am having an error when I try to run ambientContribMaximum and am unsure whether I have misunderstood its application or if I could have come across an error.
I have multiplexed data with single cells from n samples in a single raw matrix. I calculated the ambient contamination with ambientProfileEmpty(raw_matrix, good.turing = TRUE), generating a vector. I wanted to estimate the ambient contamination for all my multiplexed samples, and therefore tried running ambientContribMaximum(samples, ambient, mode="scale") where samples is a matrix of n samples. I assumed I could estimate the ambient contamination for multiple samples or clusters within a run given this line in the documentation of ambientContribMaximum: "Hence, the function works best when y contains aggregated counts for one or more groups of droplets with the same expected expression profile, e.g., clusters of related cells."
However I receive the following error when I try to run this:
Error: BiocParallel errors
element index: 1
first error: 'to' must be a finite number
In addition: Warning message:
In .maximum_ambience(y = y[, i], ambient = A, ...) :
'y' and 'ambient' do not have the same feature names
I can run it if I ensure ambient has the same number of columns as y (by copying ambient multiple times in a matrix), but I am wary about whether this is the most appropriate course of action. Have I misunderstood the application of ambientContribMaximum/is there anything I should do differently? Do you have any suggestions as to how to accomplish my goal?
Thank you in advance!
The text was updated successfully, but these errors were encountered: