Can't complete vignette start-to-finish

[mtietjen27]

Hi @warrenmcg, Me again ;-). I'm now getting a different error when I try to make the compositional sleuth object. It seems like the variable 'target_mapping' is not working eventhough 'yeastAnnos' was imported correctly.

alr_so <- make_lr_sleuth_object(sample_to_covariates = yeastS2C,
                                target_mapping = yeastAnnos, aggregation_column = 'gene_id',
                                full_model = ~condition, null_model = ~1,
                                beta = "conditionnoN", denom_name = denom)

Error in colnames(target_mapping) : object 'target_mapping' not found

Thank you!

[warrenmcg]

Hi @mtietjen27,

I again really appreciate your being an alpha tester! This patience will eventually yield you a complete analysis start-to-finish! Found the bug that caused this particular issue, and it has been pushed to master, so if you re-install, this particular error should be resolved.

[warrenmcg]

Hold off on doing anything, as I have found some other bugs that will prevent you from finishing an analysis; I have no idea how I missed this before.

[warrenmcg]

Hi @mtietjen27,

Ok, I have the vignette working on my end. Do you mind re-installing the master branch of sleuth-ALR and trying again?

[warrenmcg]

Hi @mtietjen27,

I identified a few more issues, and have fixed all of them. Please let me know if you experience any further problems on your side once you have re-installed.

Thanks!

[mtietjen27]

Hi @warrenmcg , It's moving forward but I now encounter the following error:

> alr_so <- make_lr_sleuth_object(sample_to_covariates = yeastS2C,
+                                 target_mapping = yeastAnnos, beta = 'conditionnoN',
+                                 full_model = ~condition, null_model = ~1,
+                                 denom_name = denom, run_models = TRUE)
>> 2019-03-19 11:25:13 - preparing sleuth object using the all-in-one transformation approach
reading in kallisto results
dropping unused factor levels
....
normalizing est_counts
6334 targets passed the filter
normalizing tpm
merging in metadata
summarizing bootstraps
....
Error in get_denom_names(obj) : 
  This sleuth object was not processed using compositional data analysis.
Thus its denominator does not exist.
In addition: Warning messages:
1: In (function (type = "alr", denom_name = NULL, lr_method = "both",  :
  The version of sleuth loaded is '0.30.0'. The lr_method 'both' only works with the API of sleuth in versions > 0.30.0. Setting the 'lr_method' to 'transform'...
2: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 41 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'
3: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 41 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'
4: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 32 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'
5: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 32 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'
6: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 3 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'
7: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 3 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'
8: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 9 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'
9: In impute_zeros(mat, method = impute_method, delta = delta, impute_proportion = impute_proportion) :
  There were 9 targets that had all zero values. Those targets should be removed using 'remove_essential_zeros'

Thank you!

[warrenmcg]

Hi @mtietjen27,

I obviously wasn't thorough in my testing yesterday. Thanks for your continued patience and perseverance! I resolved the new problem you uncovered. Let me know if you get any more!

[mtietjen27]

Hi @warrenmcg , It seems like you could fix all bugs - good job! And thanks a lot for solving all errors so quickly - I'm closing the issue now :-)

[warrenmcg]

Awesome news! Hope this tool helps with your analyses now that all of the bugs have been squashed.