NegProbe in Spatial Deconvolution

[artfuldodger08]

I'm using the WTA data, so I have a large number of NegProbe-WTX in the dataset. Do I need to rename each variable here for the data in order to use the Spatial Deconvolution function? When I try and do the rmNegprobe, I get errors due to the multiple rows with the same name.

And thank you for putting this together; super awesome pipeline to use.

[ningbioinfo]

Hi @artfuldodger08 , for multiple negprobe please refer to this post: #33

There is a quick fix by simply re-labeling the name, such as adding number to the back of the string. For me I would do it before readGeoMx so read the count table in as a dataframe, mutate the gene name column, something like

df <- df %>%
    mutate(negprobe_count = cumsum(grepl("NegProbe",genecol)),  
           genecol = if_else(grepl("NegProbe",genecol), 
                          paste0(genecol, negprobe_count), 
                          genecol)) %>%
    select(-negprobe_count) 

PS: didn't test it may have bugs.

Then when prepare for the deconvolution make use of the pool argument like I mention in the previous post.

PPS: I swear I'll make an update on this by detecting multi negative probes and rename it automatically...