Small Issue - Hard Coded timepoint metadata in get_perturbation_effects()

[acmullen-med]

In get_perturbation_effects() on the output of fit_genotype_ccm() when passing interval_col = "timepointNumeric" I was getting a crash. When I renamed the colData(cds) to timepoint, remade ccs, remade ccm, it didn't crash.

Original

> testFitGenotypeCCM <- fit_genotype_ccm(genotype="Myod1Myf5", ccs=ccsPerturbV3Partition, interval_col="timepointNumeric", perturbation_col="perturbation",ctrl_ids=c("Control"),num_threads=25,num_bootstraps=10,penalize_by_distance = FALSE)
Fitting knockout model for Myod1Myf5
        time range: 11 to 48
        formula: ~  ns( timepointNumeric , knots= c(29.5) ) +  ns( timepointNumeric , knots= c(29.5) ):knockout + knockout
[1] "fitting model with 25 threads"
Warning message:
In Robject$getModel(var, index) :
  No such a model in the collection. Acceptable parameter values can be found via
$ranks (for PCA)
$clusters (for mixture models)
$penalties (for network)
Returning model with closest value. Requested: 0.000294161464618765 , returned: 0.00271708372175912
> perturb_effects <- platt::get_perturbation_effects(testFitGenotypeCCM,
  interval_col = "timepointNumeric",
  adjust_q_values = FALSE
)
Error in `mutate()`:
ℹ In argument: `genotype_eff = purrr::map(.f = make_contrast, .x = data, ccm = ccm, adjust_q_values = adjust_q_values)`.
Caused by error in `purrr::map()`:
ℹ In index: 1.
Caused by error in `mutate()`:
ℹ In argument: `timepoint_abund = purrr::map(...)`.
Caused by error in `purrr::map()`:
ℹ In index: 1.
Caused by error:
! object 'timepointNumeric' not found
Run `rlang::last_trace()` to see where the error occurred.
> 

Full Stack Trace for debugging.

> rlang::last_trace()
<error/dplyr:::mutate_error>
Error in `mutate()`:
ℹ In argument: `genotype_eff = purrr::map(.f = make_contrast, .x = data, ccm = ccm, adjust_q_values = adjust_q_values)`.
Caused by error in `purrr::map()`:
ℹ In index: 1.
Caused by error in `mutate()`:
ℹ In argument: `timepoint_abund = purrr::map(...)`.
---
Backtrace:
     ▆
  1. ├─platt::get_perturbation_effects(...)
  2. │ └─... %>% unnest(c(data, genotype_eff))
  3. ├─tidyr::unnest(., c(data, genotype_eff))
  4. ├─dplyr::mutate(...)
  5. ├─dplyr:::mutate.data.frame(...)
  6. │ └─dplyr:::mutate_cols(.data, dplyr_quosures(...), by)
  7. │   ├─base::withCallingHandlers(...)
  8. │   └─dplyr:::mutate_col(dots[[i]], data, mask, new_columns)
  9. │     └─mask$eval_all_mutate(quo)
 10. │       └─dplyr (local) eval()
 11. ├─purrr::map(.f = make_contrast, .x = data, ccm = ccm, adjust_q_values = adjust_q_values)
 12. │ └─purrr:::map_("list", .x, .f, ..., .progress = .progress)
 13. │   ├─purrr:::with_indexed_errors(...)
 14. │   │ └─base::withCallingHandlers(...)
 15. │   ├─purrr:::call_with_cleanup(...)
 16. │   └─platt (local) .f(.x[[i]], ...)
 17. │     └─hooke::estimate_abundances(ccm, newdata = newdata_wt)
 18. │       └─... %>% tidyr::unnest(c(timepoint_abund))
 19. ├─tidyr::unnest(., c(timepoint_abund))
 20. ├─dplyr::select(., timepoint_abund)
 21. ├─dplyr::mutate(...)
 22. ├─dplyr:::mutate.data.frame(...)
 23. │ └─dplyr:::mutate_cols(.data, dplyr_quosures(...), by)
 24. │   ├─base::withCallingHandlers(...)
 25. │   └─dplyr:::mutate_col(dots[[i]], data, mask, new_columns)
 26. │     └─mask$eval_all_mutate(quo)
 27. │       └─dplyr (local) eval()
 28. ├─purrr::map(...)
 29. │ └─purrr:::map_("list", .x, .f, ..., .progress = .progress)
 30. │   ├─purrr:::with_indexed_errors(...)
 31. │   │ └─base::withCallingHandlers(...)
 32. │   ├─purrr:::call_with_cleanup(...)
 33. │   └─hooke (local) .f(.x[[i]], ...)
 34. │     └─Matrix::sparse.model.matrix(model_terms, newdata, xlev = ccm@model_aux[["full_model_xlevels"]])
 35. │       ├─stats::model.frame(object, data, xlev = xlev)
 36. │       └─stats::model.frame.default(object, data, xlev = xlev)
 37. │         └─base::eval(predvars, data, env)
 38. │           └─base::eval(predvars, data, env)
 39. │             └─splines::ns(...)
 40. ├─base::.handleSimpleError(...)
 41. │ └─purrr (local) h(simpleError(msg, call))
 42. │   └─cli::cli_abort(...)
 43. │     └─rlang::abort(...)
 44. │       └─rlang:::signal_abort(cnd, .file)
 45. │         └─base::signalCondition(cnd)
 46. └─dplyr (local) `<fn>`(`<prrr_rr_>`)
---
Caused by error in `purrr::map()`:
ℹ In index: 1.
Caused by error:
! object 'timepointNumeric' not found
---
Backtrace:
     ▆
  1. ├─platt::get_perturbation_effects(...)
  2. │ └─... %>% unnest(c(data, genotype_eff))
  3. ├─tidyr::unnest(., c(data, genotype_eff))
  4. ├─dplyr::mutate(...)
  5. ├─dplyr:::mutate.data.frame(...)
  6. │ └─dplyr:::mutate_cols(.data, dplyr_quosures(...), by)
  7. │   ├─base::withCallingHandlers(...)
  8. │   └─dplyr:::mutate_col(dots[[i]], data, mask, new_columns)
  9. │     └─mask$eval_all_mutate(quo)
 10. │       └─dplyr (local) eval()
 11. ├─purrr::map(.f = make_contrast, .x = data, ccm = ccm, adjust_q_values = adjust_q_values)
 12. │ └─purrr:::map_("list", .x, .f, ..., .progress = .progress)
 13. │   ├─purrr:::with_indexed_errors(...)
 14. │   │ └─base::withCallingHandlers(...)
 15. │   ├─purrr:::call_with_cleanup(...)
 16. │   └─platt (local) .f(.x[[i]], ...)
 17. │     └─hooke::estimate_abundances(ccm, newdata = newdata_wt)
 18. │       └─... %>% tidyr::unnest(c(timepoint_abund))
 19. ├─tidyr::unnest(., c(timepoint_abund))
 20. ├─dplyr::select(., timepoint_abund)
 21. ├─dplyr::mutate(...)
 22. ├─dplyr:::mutate.data.frame(...)
 23. │ └─dplyr:::mutate_cols(.data, dplyr_quosures(...), by)
 24. │   ├─base::withCallingHandlers(...)
 25. │   └─dplyr:::mutate_col(dots[[i]], data, mask, new_columns)
 26. │     └─mask$eval_all_mutate(quo)
 27. │       └─dplyr (local) eval()
 28. └─purrr::map(...)
 29.   └─purrr:::map_("list", .x, .f, ..., .progress = .progress)
 30.     ├─purrr:::with_indexed_errors(...)
 31.     │ └─base::withCallingHandlers(...)
 32.     ├─purrr:::call_with_cleanup(...)
 33.     └─hooke (local) .f(.x[[i]], ...)
 34.       └─Matrix::sparse.model.matrix(model_terms, newdata, xlev = ccm@model_aux[["full_model_xlevels"]])
 35.         ├─stats::model.frame(object, data, xlev = xlev)
 36.         └─stats::model.frame.default(object, data, xlev = xlev)
 37.           └─base::eval(predvars, data, env)
 38.             └─base::eval(predvars, data, env)
 39.               └─splines::ns(...)

Change to
testFitGenotypeCCM <- fit_genotype_ccm(genotype="Myod1Myf5", ccs=ccsPerturbV3Partition, interval_col="timepoint", perturbation_col="perturbation",ctrl_ids=c("Control"),num_threads=25,num_bootstraps=10,penalize_by_distance = FALSE)

Confirming that the information in the colData is the same.

> dim(cdsPerturbV3rowDataSubset)
sum(cdsPerturbV3rowDataSubset$timepointNumeric == cdsPerturbV3rowDataSubset$timepoint)
[1]  30000 272738
[1] 272738