plot_volcano: Object 'b' not found

[umasstr]

I assume that this error means that plot_volcano cannot find my Wald beta values? I believe that I ran sleuth_wt successfully. Any help would be greatly appreciated:

models(so) [ full ] formula: ~condition data modeled: obs_counts transform sync'ed: TRUE coefficients: (Intercept) conditionok186 [ reduced ] formula: ~1 data modeled: obs_counts transform sync'ed: TRUE coefficients: (Intercept)

tests(so) ~likelihood ratio tests: reduced:full ~wald tests: [ full ] conditionok186

plot_volcano(so,"conditionok186",test_type = "wt") Error in FUN(X[[i]], ...) : object 'b' not found

[pimentel]

hm, it seems to not be pulling out the effect sizes from the sleuth_results table.

can you do me a favor?

debugonce(plot_volcano)
plot_volcano(so,"conditionok186",test_type = "wt")

then hit enter until you get to the line that says:

  res <- dplyr::mutate(res, significant = qval < sig_level)

then report back the result from:

colnames(res)

thanks in advance

[Rhinogradentia]

This seems to be the same error as in #233 - there it is a missing parameter for sleuth_results.

[steachhr]

Hi,

I'm getting the same 'object b not found' error while trying to generate volcano plots. I ran the code above, and the colnames(res) returns:

[1] "target_id" "ext_gene"
[3] "num_aggregated_transcripts" "sum_mean_obs_counts"
[5] "pval" "qval"

[umasstr]

Hi all, there may have been a problem with gene/tx aggregation in sleuth prep. I used an updated biomart ensembl host and my issue was resolved. Not sure if this is related, but hope it helps...

[ymatmt]

Hi,

I used an updated biomart ensembl host and my issue was resolved.

How do you do it? I got a same error.

[umasstr]

Hi ymatmt, I stopped using an old ensembl archive ("archive.dec2015...") in the mart variable, and instead use host = "www.ensembl.org"). I'm not sure why this helped. I'd be happy to test out a couple of your abundance files if you'd like.

[ymatmt]

Finally it works! Thank you!