Possible check in liana_prep

[welch16]

Hello,

Thanks for sharing this great package. I was trying to use liana to analyze my data, which was a SingleCellExperiment object that I have been using for a while.

When I tried to use liana_wrap directly, I found the following error:

Error in `map()`:
ℹ In index: 1.
ℹ With name: CellChatDB.
Caused by error in `map()`:
ℹ In index: 1.
Caused by error in `h()`:
! error in evaluating the argument 'x' in selecting a method for function 'calculateAverage': level sets of factors are different
Run `rlang::last_trace()` to see where the error occurred.

which wasn't really informative, using rlang::last_trace didn't revealed a lot either at first glance:

  1. ├─liana::liana_wrap(...)
  2. │ └─... %>% setNames(names(resource))
  3. ├─stats::setNames(., names(resource))
  4. ├─purrr::map(...)
  5. │ └─purrr:::map_("list", .x, .f, ..., .progress = .progress)
  6. │   ├─purrr:::with_indexed_errors(...)
  7. │   │ └─base::withCallingHandlers(...)
  8. │   ├─purrr:::call_with_cleanup(...)
  9. │   └─liana (local) .f(.x[[i]], ...)
 10. │     └─rlang::invoke(...)
 11. ├─liana (local) .fn(...)
 12. │ └─liana:::get_log2FC(sce, assay.type = assay.type, base)
 13. │   └─... %>% unnest(value)
 14. ├─tidyr::unnest(., value)
 15. ├─tibble::enframe(., name = "cell")
 16. ├─stats::setNames(., levels(colLabels(sce)))
 17. ├─purrr::map(...)
 18. │ └─purrr:::map_("list", .x, .f, ..., .progress = .progress)
 19. │   ├─purrr:::with_indexed_errors(...)
 20. │   │ └─base::withCallingHandlers(...)
 21. │   ├─purrr:::call_with_cleanup(...)
 22. │   └─liana (local) .f(.x[[i]], ...)
 23. │     ├─... %>% dplyr::rename(subject_avg = value)
 24. │     ├─scater::calculateAverage(...)
 25. │     ├─base::subset(sce, select = colLabels(sce) == subject)
 26. │     └─S4Vectors::subset.Vector(sce, select = colLabels(sce) == subject)
 27. │       ├─base::subset(x, ...)
 28. │       └─SummarizedExperiment::subset(x, ...)
 29. │         └─SummarizedExperiment (local) .local(x, ...)
 30. │           └─S4Vectors:::evalqForSubset(select, colData(x), ...)
 31. │             └─S4Vectors:::evalArg(substitute(expr), envir, ..., where = parent.frame())
 32. │               └─S4Vectors:::safeEval(expr, envir, enclos, ...)
 33. │                 ├─BiocGenerics::eval(expr, envir, enclos)
 34. │                 └─S4Vectors::eval(expr, envir, enclos)
 35. │                   ├─BiocGenerics::eval(expr, as.env(envir, enclos))
 36. │                   └─base::eval(expr, as.env(envir, enclos))
 37. │                     └─base::eval(expr, as.env(envir, enclos))
 38. │                       └─base::Ops.factor(colLabels(sce), subject)
 39. │                         └─base::stop("level sets of factors are different")
 40. ├─dplyr::rename(., subject_avg = value)
 41. ├─tibble::as_tibble(., rownames = "gene")
 42. └─base::.handleSimpleError(...)
 43.   └─base (local) h(simpleError(msg, call))

But it turned out that the reason that it was crashing was in the line S4Vectors::subset.Vector(sce, select = colLabels(sce) == subject) because I had a subject column in my colData. After changing the name of the column it worked fine, so I thought to let you know about this error because it can be easily addressed in the liana_prep function.

Thanks again, Rene

[dbdimitrov]

Thanks Rene @welch16

I will make sure to change this in the next update :)