plot_cpdb() error

[rosanababu]

Hi,

Many thanks for the amazing tool. I'm using ktplots and cellphoneDB for the first time. I've used cpdb_statistical_analysis_method from latest cellphoneDB v5 to run statistical analysis. plot_cpdb_heatmap() works fine on the results, unfortunately dot plot plot_cpdb() is showing error. Please could you help to solve the following error from plot_cpdb().

Dot plot

plot_cpdb(

• scdata=sce, 
  

• cell_type1="Pericyte",
  

• cell_type2=".",  
  

• celltype_key="Cell_type",
  

• means=means_stat,
  

• pvals=pvals_stat,
  

• genes=c("TNFRSF13B", "STAT5A"),
  

• title="interacting interactions!",
  

• )
  Error in dplyr::left_join():
  ! Can't join x$Var2 with y$Var2 due to incompatible types.
  ℹ x$Var2 is a .
  ℹ y$Var2 is a <factor<4887b>>.
  Run rlang::last_trace() to see where the error occurred.

rlang::last_trace()
<error/dplyr_error_join_incompatible_type>
Error in dplyr::left_join():
! Can't join x$Var2 with y$Var2 due to incompatible types.
ℹ x$Var2 is a .
ℹ y$Var2 is a <factor<4887b>>.

---

Backtrace:
▆

1. └─ktplots::plot_cpdb(...)
2. ├─dplyr::left_join(df_means, df_pvals, by = c("Var1", "Var2"))
3. └─dplyr:::left_join.data.frame(df_means, df_pvals, by = c("Var1", "Var2"))
   Run rlang::last_trace(drop = FALSE) to see 22 hidden frames.

rlang::last_trace(drop = FALSE)
<error/dplyr_error_join_incompatible_type>
Error in dplyr::left_join():
! Can't join x$Var2 with y$Var2 due to incompatible types.
ℹ x$Var2 is a .
ℹ y$Var2 is a <factor<4887b>>.

---

Backtrace:
▆

1. ├─ktplots::plot_cpdb(...)
2. │ ├─dplyr::left_join(df_means, df_pvals, by = c("Var1", "Var2"))
3. │ └─dplyr:::left_join.data.frame(df_means, df_pvals, by = c("Var1", "Var2"))
4. │ └─dplyr:::join_mutate(...)
5. │ └─dplyr:::join_cast_common(x_key, y_key, vars, error_call = error_call)
6. │ ├─rlang::try_fetch(...)
7. │ │ └─base::withCallingHandlers(...)
8. │ └─vctrs::vec_ptype2(x, y, x_arg = "", y_arg = "", call = error_call)
9. ├─vctrs (local) <fn>()
10. │ └─vctrs::vec_default_ptype2(...)
11. │ ├─base::withRestarts(...)
12. │ │ └─base (local) withOneRestart(expr, restarts[[1L]])
13. │ │ └─base (local) doWithOneRestart(return(expr), restart)
14. │ └─vctrs::stop_incompatible_type(...)
15. │ └─vctrs:::stop_incompatible(...)
16. │ └─vctrs:::stop_vctrs(...)
17. │ └─rlang::abort(message, class = c(class, "vctrs_error"), ..., call = call)
18. │ └─rlang:::signal_abort(cnd, .file)
19. │ └─base::signalCondition(cnd)
20. └─rlang (local) <fn>(<vctrs__2>)
21. └─handlers[1L]

22. └─dplyr:::rethrow_error_join_incompatible_type(cnd, vars, error_call)
    

23.   └─dplyr:::stop_join(...)
    

24.     └─dplyr:::stop_dplyr(...)
    

25.       └─rlang::abort(...)
    

Kind regards,
RB

[zktuong]

hi @rosanababu , can i ask you to check if the column names for both your means_stat and pvals_stat are similar?

Also are the dimensions the same/similar?

[biobirds9]

hi @rosanababu , can i ask you to check if the column names for both your and are similar?means_stat``pvals_stat

Also are the dimensions the same/similar?

Hi, I have also encountered this problem and I have confirmed that the column names of both files are exactly the same, how can I fix this error?

[zktuong]

what if you remove the genes option?

[rosanababu]

hi @rosanababu , can i ask you to check if the column names for both your means_stat and pvals_stat are similar?

Also are the dimensions the same/similar?

Hi, Thanks for your reply. Yes, the dimension and the column names for both means_stat and ovals_stat are similar. I could run it without any errors when I filtered out cells that has no UMIs using: seurat_subset <- subset(seurat, nCount_RNA > 10). Then, I created the metadata and counts matrices from seurat_subset to input CellPhoneDB. Now the outputs of CellphoneDB pvals_stat and means_stat is working fine even when I set splitkey for conditions. Thank you

[zktuong]

no more issues so i'm going to assume it working now.