double conditions in source code not supported by R anymore?

[Taib98]

Hello,

While trying the toy dataset analysis based on the tutorial, I consistently got an error when trying to extract the mutual information and entropy information. There is apparently a double condition (i.e &&) that R signals as an error in the source code. Is there any way to fix this issue?

for (i in 1:length(s)){
  ent_mi_data <- calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor = 1.5, bin_size = 0.1)
  #calculate entropy and mutual  info in first andd second element based on treatment
  s[[i]]$mutual_inf <- ent_mi_data[["data_dists"]][['MI']]
  s[[i]]$entropy <- ent_mi_data[["data_dists"]][["ENT"]]
}
#gives error: Error in well_i != well_j && same_well_only == T : 'length = 2' in coercion to 'logical(1)'

Thanks a lot,
Kind regards,

T.I.

[sje30]

hi,
thanks for your issue. I'll try to help navigate it. Can you just point me to the code that you were running, and state how you ran it?

Was it here:

meaRtools/MEA_analysis/Code/IGM_main.R

Line 93 in 0b04c9c

ent_mi_data <- calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor=1.5, bin_size=0.1)

and if so, what data file did you provide?

[Taib98]

Hi,

Thank you for your reaction! Yes, the function I used was this one. I followed the steps written in the vignette (https://github.com/igm-team/meaRtools/blob/master/meaRtools/vignettes/meaRtoolsGeneralUsage.Rmd). In case you need more information, here is the code I ran up until the error:

spk_list_files <- c(system.file("extdata","exampleRecording_1012016_plate1_DIV3_spike_list.csv.gz",
                                package = "meaRtools"),
                    system.file("extdata","exampleRecording_1012016_plate1_DIV4_spike_list.csv.gz",
                                package = "meaRtools"))

experimental_log_file <- system.file("extdata","exampleRecording_1012016_plate1_expLog.csv.gz",
                                     package = "meaRtools")

data_dir<-dirname(spk_list_files[1]) 

output_dir<-paste0( data_dir , "/Analysis" ) 

suppressWarnings(  dir.create(paste(getwd(),output_dir, sep = "") ) 

output_perDIV_dir<-paste0( data_dir , "/Analysis/outputPerDIV" ) 
suppressWarnings(  dir.create(output_perDIV_dir) ) 


r_object_dir<-paste0( data_dir , "/Analysis/R_Objects" )
suppressWarnings(  dir.create(r_object_dir) )

log.dir<-paste0( output_dir , "/LogFiles" ) 
suppressWarnings(  dir.create(log.dir) )

analysis<-list(spikeFiles = spk_list_files, output_dir = output_dir,
               r_output_dir = r_object_dir, output_perDIV_dir = output_perDIV_dir)

for (i in 1:length(spk_list_files)){
  title<-strsplit(basename(spk_list_files[i]), ".csv")[[1]][1]
  plate_chem_info<-get_experimental_log_file(spk_list_files[i], experimental_log_file)
  analysis$Robject[i]<-read_spikelist(key=title, spk_list_file=spk_list_files[i],          chem_info=plate_chem_info,r_object_dir=r_object_dir) 
}

data("parameters") 

s<-calculate_spike_features(analysis$Robject, parameters) 

#The part giving an error:

for (i in 1:length(s)){
  ent_mi_data <- calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor=1.5, bin_size=0.1)
  #ent_mi_data <- calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor = 1.5, bin_size = 0.1)
  #calculate entropy and mutual  info in first andd second element based on treatment
  s[[i]]$mutual_inf <- ent_mi_data[["data_dists"]][['MI']]
  s[[i]]$entropy <- ent_mi_data[["data_dists"]][["ENT"]]
}
#gives error: Error in well_i != well_j && same_well_only == T : 'length = 2' in coercion to 'logical(1)'

I am working on R version 4.3.1 (2023-06-16 ucrt)
The datafiles I used were the example files loaded on the github (see spk_list_files in the code chunk above).

Best,
T.I.

[Taib98]

Hi,

I am reaching out to this issue again, as I still haven't found a solution to this problem. Do you think you could help me solve it ? Let me know if any additional information would be required.

Best,
T.I.

[sje30]

I'm very sorry for the slowness of my reply. This is now software that I do not actively use currently, and so will need to reinstall dependencies. In the meantime, please can you do the following when you get the error:

traceback()

[Taib98]

Thank you for your reaction, here is what I get with traceback()

> s<-calculate_spike_features(analysis$Robject, parameters) 
> for (i in 1:length(s)){
+   ent_mi_data <- calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor=1.5, bin_size=0.1)
+   #ent_mi_data <- calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor = 1.5, bin_size = 0.1)
+   #calculate entropy and mutual  info in first andd second element based on treatment
+   s[[i]]$mutual_inf <- ent_mi_data[["data_dists"]][['MI']]
+   s[[i]]$entropy <- ent_mi_data[["data_dists"]][["ENT"]]
+ }
Error in well_i != well_j && same_well_only == T : 
  'length = 2' in coercion to 'logical(1)'

> traceback()
3: .electrode_dist_pairs(mea, elec_names = well_elec_names, dist_max = dist_max)
2: .pairwise_dists_per_well(mea, wellnames = treatment_wells, dist_metric = "mutual_information", 
       bin_size)
1: calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor = 1.5, 
       bin_size = 0.1)

Best,
T.I.

[spavlovska]

Hello,

I am also trying to implement your code for analyzing my MEA recordings and unfortunately have encountered the same issue with "calculate_entropy_and_mi" function. I tried it with the example recordings from your github and also with my own data, but get the same error as T.I with both data sets.

# As mutual information and entropy values are considered spike features, we will calculate them here > # based on the spike data of each electrode > for (i in 1:length(s)) { + ent_mi_data <- calculate_entropy_and_mi(s[[i]], s[[i]]$treatment, mult_factor=1.5, bin_size=0.1) + s[[i]]$mutual_inf <- ent_mi_data[["data_dists"]][["MI"]] + s[[i]]$entropy <- ent_mi_data [["data_dists"]][["ENT"]] + } Error in well_i != well_j && same_well_only == T : 'length = 2' in coercion to 'logical(1)' --   > | > >

I was wondering if you had any updates for the issue or found a way around it. Thank you very much for your help, time, and developing the tool.

Best,
SP