About checked$outliers about mdseq

[zjdaye]

Hi Kevin,

Thanks for your interest!

It seems that an error is encountered in the process, so the detection stopped and no outlier detection was performed. (A non-zero status indicates an error, and the num.outliers is NA or not available.) Usually, this might be encountered on only a few unusually genes when the data cannot be properly estimated; for example, if all counts were zero. Please see Supplementary materials for more detailed discussions.

I suggest that you check your data first to see if there might be anything peculiar and compare with the vignette. You can also try outlier check on individual genes with check.outlier function (setting verbose=TRUE to get more info). Finally, you can try glm.MD function to see if the counts data were properly fitted.

Please let us know how it goes! If necessary for us to look in more details, please send us an example data and your codes/outputs. (Can send to my email address in the paper.)

Good luck!

Best,
John.