error from make_brainGraphList

[mariaeugeni...@gmail.com]

Dear Chris,

I have recently tried to perform an analysis on probtrackx data using R v. 4.0.2 and brainGraph 3.0.1 on a macOS Mojave 10.14.6.

It seems that my data is uploaded correctly, both the matrices and my custom atlas (I already used a custom atlas with previous versions and it worked fine). When I run the create_mats command, however, I get this error:

(with j=1)

> g_l[[j]] <- make_brainGraphList(A.norm.sub_l[[j]], atlas=atlas_bil, modality=modality, weighting='sld', threshold=thresholds[j], weighted=TRUE, gnames=covars.dti$Study.ID, groups=covars.dti$Group)

  |                                                                                                           |   0%Error in { : task 1 failed - "Invalid first argument"

This is the traceback output:

> traceback()

5: stop(simpleError(msg, call = expr))

4: e$fun(obj, substitute(ex), parent.frame(), e$data)

3: foreach(i = seq_len(kNumGraphs)) %dopar% {

       res <- loopfun(x[, , i], atlas, type, level, set.attrs, modality, 

           weighting, threshold[i], name = gnames[i], Group = grpNames[i], 

           subnet = subnet, mode = mode, diag = diag, weighted = weighted, 

           use.parallel = FALSE, ...)

   }

2: make_brainGraphList.array(A.norm.sub_l[[j]], atlas = atlas_bil, 

       modality = modality, weighting = "sld", threshold = thresholds[j], 

       weighted = TRUE, gnames = covars.dti$Study.ID, groups = covars.dti$Group)

1: make_brainGraphList(A.norm.sub_l[[j]], atlas = atlas_bil, modality = modality, 

       weighting = "sld", threshold = thresholds[j], weighted = TRUE, 

       gnames = covars.dti$Study.ID, groups = covars.dti$Group)

I have tried the script also on a Ubuntu 16 machine, but I get the same error. Do you have any ideas on how to solve it?

Another thing: I want to normalize my matrices using rowSums. What files do I need to specify to div.files in create_mats?

All the best,

Maria Eugenia

[Chris Watson]

1. Even though the body of your email says the error occurs with "create_mats", it looks like it occurs with "make_brainGraphList". I cannot diagnose the issue without seeing the data myself, but it is possible there is a subject with negative weights that is causing the error. You could try running "make_brainGraph" on the first subject and see if that runs without error.

2. You don't need any additional files to normalize by row sums; it calculates the row sums of the input matrices.

Chris

--
You received this message because you are subscribed to the Google Groups "brainGraph-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brainGraph-he...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/9b4cb8a2-9ae2-4b98-87ad-c419d88302a0n%40googlegroups.com.

[Femke Lammertink]

Hi Maria,

I'm having a similar issue, same traceback: task 3 failed - "'x' must be an array of at least two dimensions"

5: stop(simpleError(msg, call = expr))
4: e$fun(obj, substitute(ex), parent.frame(), e$data)
3: foreach(i = seq_len(kNumGraphs)) %dopar% {
       res <- loopfun(x[, , i], atlas, type, level, set.attrs, modality,
           weighting, threshold[i], name = gnames[i], Group = grpNames[i],
           subnet = subnet, mode = mode, diag = diag, weighted = weighted,
           use.parallel = FALSE, ...)
   }

2: make_brainGraphList.array(my.mats$A.norm.sub[[j]], atlas = atlas,
       modality = modality, weighting = "fa", weighted = TRUE, threshold = thresholds[j],
       gnames = Temp$SubjectID, grpNames = Temp$SessionID)
1: make_brainGraphList(my.mats$A.norm.sub[[j]], atlas = atlas, modality = modality,
       weighting = "fa", weighted = TRUE, threshold = thresholds[j],
       gnames = Temp$SubjectID, grpNames = Temp$SessionID)

I checked "x[,,i]" and it is actually able to find the data...

Where you able to solve it?

Many thanks!

Femke

[Chris Watson]

Hi Femke, as I previously told Maria, you should try running "make_brainGraph" on a single subject first. The traceback from errors occurring in a "foreach" loop are rarely informative so both of you may be seeing completely different errors.

Chris

--

You received this message because you are subscribed to the Google Groups "brainGraph-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brainGraph-he...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/f1143861-7d5c-46e6-9a60-01da16af57c2n%40googlegroups.com.

[Femke Lammertink]

Dear Chris,

My apologies. I ran the single subject, and got the following traceback:

Error in seq_len(no) : argument must be coercible to non-negative integer

in addition: Warning message:

In vattrs[[name]][index] <- value :

number of items to replace is not a multiple of replacement length

17: eval(x$expr, data, x$env)
16: eval(x$expr, data, x$env)
15: FUN(X[[i]], ...)
14: lapply(x, lazy_eval, data = data)
13: lazy_eval(args, data = c(attrs, .inc = .inc, inc = inc, adj = adj,
        .from = .from, from = from, .to = .to, to = to, .igraph.from = list(.Call(C_R_igraph_mybracket,
            graph, 3L)[as.numeric(x)]), .igraph.to = list(.Call(C_R_igraph_mybracket,
            graph, 4L)[as.numeric(x)]), .igraph.graph = list(graph),
        `%--%` = `%--%`, `%->%` = `%->%`, `%<-%` = `%<-%`))
12: `[.igraph.es`(edges, seq_len(no))
11: edges[seq_len(no)]
10: na.omit(es)
9: as.igraph.es(graph, na.omit(es))
8: ends(x, edges[seq_len(no)], names = names)
7: x("width", no = can_max)
6: head_print_callback(x, max_lines, header, footer, omitted_footer,
       ...)
5: head_print(fun, max_lines = max.lines)
4: .print.edges.compressed.limit(x, edges, names, max.lines)
3: .print.edges.compressed(x, names = names, max.lines = max.lines -
       head_lines)
2: print.igraph(x)
1: (function (x, ...)
   UseMethod("print"))(x)

## Code below generated same error

make_brainGraph(A.norm.sub[[1]][, , 2], atlas='custom_atlas', modality=modality,
                weighting='fa', threshold=thresholds[1],
                weighted=TRUE, name=Temp$SubjectID[1], Group=Temp$SessionID[1])

g.tmp <- graph_from_adjacency_matrix(A.norm.sub[[1]][, , 1], mode='undirected', diag=FALSE, weighted=TRUE)
make_brainGraph(g.tmp, atlas='custom_atlas', modality=modality,
                weighting='fa', threshold=thresholds[1],
                name=Temp$SubjectID[1], Group=Temp$SessionID[1])

I attached the R environment (subset of sample).

Thought the issue was related to the custom atlas, but now I'm not sure anymore.

[Chris Watson]

That "error" is not a real error and only relates to printing the graph information (i.e., the "print.igraph" function).

I was able to successfully run "make_brainGraphList" across all your thresholds so I am not sure why you are getting an error.

Chris

--

You received this message because you are subscribed to the Google Groups "brainGraph-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brainGraph-he...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/7a980ab5-f49e-49a6-9b0b-14aa7e9b52dbn%40googlegroups.com.

[Femke Lammertink]

Dear Chris,

Thanks for checking! Good to know that you got it working. Will spend some more time on it then :)

[Reyhaneh Bakhtiari]

Hi,

I received the same 

Error in seq_len(no) : argument must be coercible to non-negative integer

with both my own data and the attached data/code.

Any suggestion on how to overcome it?

Kind regards,

Reyhaneh

[Chris Watson]

Hi Reyhaneh, you didn't attach anything (as far as I can see), but either way that error typically only appears when printing/viewing a graph. I myself have only seen that error in those cases.

If this isn't the case, please provide more information: the code you ran, the error message, the output from "traceback()", and the output from "sessionInfo()".

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/ce5c3ec0-be4a-4a43-893b-fa38be9cdc75n%40googlegroups.com.

[Reyhaneh Bakhtiari]

Hi Chris,

I get this error at the very first stage of using BrainGraph, which is running this code:

BrainGraphmatrix<-make_brainGraphList(covars,ROI19tmp, modality='fMRI', weighting='partial', weighted=TRUE

I attached covars & ROI19tmp.

Kind regards,

Reyhaneh

[Chris Watson]

The "make_brainGraphList" command should not be the first that you run (in nearly every situation).

Please read through the User Guide to see how to start analyzing your data. But here are a few comments:

1. In particular, the variable you have named "covars" looks to be a (connectivity) matrix, not a covariates table. This is not good practice so you should use meaningful variable names.

2. Second, it is a single matrix (not a 3D array), so it is for only one subject/group (I am guessing). "make_brainGraphList" is for a set of networks, not a single one.

3. Regardless, when I run "make_brainGraph" on this data, I get an error because "covars" has column names which do not match the region names in your atlas. When I remove the column names, it runs without error and returns a single "brainGraph" object.

colnames(covars) <- NULL

4. The atlas *name* should be passed, not the variable. So it should be enclosed in (single or double) quotation marks.

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/b9cbc7a6-af0f-4963-9ee7-ba0b1827581fn%40googlegroups.com.

[kcampbell]

Hi Chris, 

I am also getting an error with make_brainGraphList, but it does not appear to be any of the errors reported in this thread. 

When I run the script (as per page 46 of the User Guide for version 3), it runs for 2-3 minutes and then throws the following error: 

Error in makePSOCKcluster(names = spec, ...): 

     Cluster setup failed. 2 of 2 workers failed to connect. 

> traceback() 

5: stop(msg)

4: makePSOCKcluster(names = spec, ...)

3: makeCluster(getOption("bg.ncpus"))

2: make_brainGraphList.array(A.norm.sub[[j]], atlas = atlas, 

     modality = modality, weighting = "sld", threshold = thresholds[j],

     weighted = TRUE, names = covar.sna$Study.ID, groups = covar.sna$Group)

1: make_brainGraphList(A.norm.sub[[j]], atlas = atlas, modality = modality, 

     weighting = "sld", threshold = thresholds[j], weighted = TRUE, 

     names = covar.sna$Study.ID, groups = covar.sna$Group)

I then ran this on a single subject, and again after 2 minutes or so, it produced a very similar error: 

> make_brainGraph(A.norm.sub[[1]][, , 1], atlas = atlas, modality = modality, weighting = "sld", threshold = thresholds[1], 

weighted = TRUE, name = covar.sna$Study.ID[1], Group = covar.sna$Group[1])

Error in makePSOCKcluster(names = spec, ...) : 

     Cluster setup failed. 2 of 2 workers failed to connect.

In addition: Warning message:

In length(vattrs[[name]] <- vc : length of NULL cannot be changed

This seems like a parallel processing issue, but I have no clue how to resolve this! Any help would be greatly appreciated. 

I'm running R version 4.1.0 and brainGraph version 3.0.2 on macOS Mojave 10.14.6.

Thanks, 

Kayleigh

[Chris Watson]

I'm not sure why that error appears even when you run "make_brainGraph" for a single subject (as that doesn't involve any parallel processing). Please see this StackOverflow post for a possible workaround; you would run that code before any other code, I believe. It seems to be a problem with R >= 4.0

https://stackoverflow.com/a/64471474/3357706

There are other solutions in the linked GitHub issue if you still run into a problem.

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/b7b7aa40-72de-483c-a3ef-7b877e869175n%40googlegroups.com.

[kcampbell]

Thanks, Chris! Adding the following to my R session appeared to get it 'unstuck':

> parallel:::setDefaultClusterOptions(setup_strategy = "sequential")

...However, I'm now getting an error related to my 'custom atlas': 

Error in { : task 1 failed - "object 'jhuNeo64' not found"

> str(jhuNeo64)

Classes ‘data.table’ and 'data.frame': 64 obs. of  8 variables:

 $ name     : chr  "Thal-L" "Thal-R" "Put-L" "Put-R" ...

 $ x.mni    : num  13 -16 29 -33 21 -26 15 -19 12 -15 ...

 $ y.mni    : num  -22 -22 -2 -2 -3 -3 5 4 26 21 ...

 $ z.mni    : num  13 13 12 12 9 9 22 23 46 49 ...

 $ lobe     : Factor w/ 8 levels "cerebellum","frontal",..: 7 7 7 7 7 7 7 7 2 2 ...

 $ hemi     : Factor w/ 2 levels "left ","right ": 1 2 1 2 1 2 1 2 1 2 ...

 $ index    : int  1 2 3 4 5 6 7 8 9 10 ...

 $ name.full: chr  "thalamus" "thalamus" "putamen" "putamen" ...

 - attr(*, ".internal.selfref")=<externalptr> 

 - attr(*, "index")= int(0) 

  ..- attr(*, "__name")= int [1:64] 33 34 57 58 7 8 61 62 29 30 ...

I've supplied this identical 'custom atlas' in brainGraph v. 2.7.3 with no problems. 

Looks like this issue has also been reported here: https://github.com/cwatson/brainGraph/issues/24

Did you end up resolving this? If this is the solution: changing the foreach line to read:

g <- foreach(i=seq_len(kNumGraphs), .export=atlas) %dopar% {

How would I access and change the function?

Thanks again, 

Kayleigh

[Chris Watson]

I think if you initialize the cluster in your own script you can make the object available. Something like

cl <- makePSOCKcluster(numCores)

clusterExport(cl, jhuNeo64)

registerDoParallel(cl)

Otherwise you would have to edit "make_brainGraphList.array" locally, which could cause other problems.

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/e8e471be-51b8-4f33-b884-3b9a3ed2af13n%40googlegroups.com.

[kcampbell]

What is the object numCores? Is this the output for detectCores() ? In my case, there are 4. 

Also, clusterExport(cl, jhuNeo64) returns this error: 

Error in get(name, envir = envir) : first argument has length > 1

Oddly enough, the error I'm receiving with make_brainGraphList is sometimes: 

Error in { : task 1 failed - "unused argument (group = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3))"

Both this error and the one on "object 'jhuNeo64' not found" produce a similar traceback(). Again, not sure why this is happening because I've had no issues with the exact same data using brainGraph version < 3.0. 

I've spent several days trying to find a work around for this parallel processing issue. The only way I could get past this step was to revert back to brainGraph version 2.7.3 on R version 3.6.1 using my 2016 MacBook Pro macOS 10.14.6... 

Any other suggestion? 

Thanks, 

Kayleigh

[özgün özalay]

Hi!

I am experiencing a similar issue with the make_brainGraphList function. I am using a custom mouse atlas (formatted with as_atlas function) and correlation matrices from CONN toolbox outputs. Everything seemed OK to me until I get this error:

modality <- 'fmri'

atlas <- 'cust.atlas'

for (j in seq_along(thresholds)) {

g[[j]] <- make_brainGraphList(A.norm.sub[[j]], atlas, modality=modality,

weighting='sld', threshold=thresholds[j],

weighted=TRUE, gnames=covars.all$Study.ID,

groups=covars.all$Group)

g.group[[j]] <- make_brainGraphList(A.norm.mean[[j]], atlas, modality=modality,

weighting='sld', threshold=thresholds[j],

weighted=TRUE, gnames=grps)

}

Error in { : 

  task 1 failed - "unused argument (groups = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1))"

I have 10 subjects and only 1 group. A.norm.subj and A.norm.mean objects look OK and contain imported correlation matrices for 10 different thresholds.

I read some posts about a parallelization problem so I double-checked registerDoMC(detectCores()) and it looks OK.

You can find covars file and R save attached. I am using Ubuntu 20.04 with R version 4.0.3 and brainGraph 3.0

Any help or tips are greatly appreciated.

Best,

./OZ

[Chris Watson]

Hi Ozgun, for your error you need to change the "groups" argument to be "grpNames".

But you will run into another error after that, a bug in "within_module_deg_z_score". Since it will take awhile for me to get a bugfix, you can try editing the file "vertex_roles.R" and sourcing it. The line to change has "rowSums" in it. It should look like (I think, I haven't tested it):

x <- rowSums(A[memb == S, memb == S, drop=FALSE])

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/27982bc6-c232-4dcd-b717-7c20a0eccc48n%40googlegroups.com.

[özgün özalay]

Thanks a lot for quick response Chris. 

Error is gone after I changed the argument name and now I am trying to edit/source vertex_roles.R file (I've never done it before, so it may take some time). But still, I have another error complaining about negative values in matrices:

Error in { : 

  task 1 failed - "At centrality.c:2058 : Weight vector must be non-negative, Invalid value"

traceback() is exactly same with the previous fixed error:

5: stop(simpleError(msg, call = expr))

4: e$fun(obj, substitute(ex), parent.frame(), e$data)

3: foreach(i = seq_len(kNumGraphs)) %dopar% {

       res <- loopfun(x[, , i], atlas, type, level, set.attrs, modality, 

           weighting, threshold[i], name = gnames[i], Group = grpNames[i], 

           subnet = subnet, mode = mode, diag = diag, weighted = weighted, 

           use.parallel = FALSE, ...)

   }

2: make_brainGraphList.array(A.norm.sub[[j]], atlas, modality = modality, 

       weighting = "sld", threshold = thresholds[j], weighted = TRUE, 

       gnames = covars.all$Study.ID, grpNames = covars.all$Group)

1: make_brainGraphList(A.norm.sub[[j]], atlas, modality = modality, 

       weighting = "sld", threshold = thresholds[j], weighted = TRUE, 

       gnames = covars.all$Study.ID, grpNames = covars.all$Group)

Should I remove negative values from the matrices? Or something else?

I will let you know about vertex_roles.R file modification.

Best,

./OZ

[Chris Watson]

The negative weights should cause the functions to skip certain calculations, but you can try removing them.

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/4df1e4eb-3225-4d3f-8eaf-24122d7d878an%40googlegroups.com.

[özgün özalay]

Removing negative values from correlation matrices worked (at least looks like) and I don't get any errors now. But now the problem is when I run;

for (j in seq_along(thresholds)) {

    g[[j]] <- make_brainGraphList(A.norm.sub[[j]], atlas='level1.atlas', modality=modality,

                                  weighting='sld', threshold=thresholds[j],

                                  weighted=TRUE, gnames=covars.all$Study.ID,

                                  grpNames=covars.all$Group)

    

    g.group[[j]] <- make_brainGraphList(A.norm.mean[[j]], atlas='level1.atlas', modality=modality,

                                        weighting='sld', threshold=thresholds[j],

                                        weighted=TRUE, gnames=grps)

}

I get several lines of output like below and it stops without an error:

[1] "Start time: 2021-07-21 17:26:18"

  |                                                                                                                       |   0%

[1] "Start time: 2021-07-21 17:26:18"

  |                                                                                                                       |   0%

[1] "Start time: 2021-07-21 17:26:18"

  |                                                                                                                       |   0%

I checked the vertex_roles.R file but the function "within_module_deg_z_score" doesn't have any lines containing "rowSums". Here you can take a look at the function:

within_module_deg_z_score <- function(g, memb, A=NULL) {

  stopifnot(is_igraph(g))

  N <- max(memb)

  if (is.null(A)) A <- as_adj(g, sparse=FALSE, names=FALSE)

  z <- Ki <- rep.int(0, dim(A)[1L])

  Ksi <- sigKsi <- rep.int(0, N)

  for (S in seq_len(N)) {

    x <- A[memb==S, ] %*% (memb==S)

    Ki[memb==S] <- x

    Ksi[S] <- mean(x)

    sigKsi[S] <- sd(x)

  }

  z <- (Ki - Ksi[memb]) / sigKsi[memb]

  z[is.infinite(z)] <- 0

  return(z)

}

 

I am trying to use mouse rsfmri data with a low number of ROIs (44) and MNI coordinates in my custom atlas may not be 100% accurate. Can any of these cause this problem?

Best,

./OZ

[Chris Watson]

That's an older version of the function. Please make sure you are using the latest version of brainGraph.

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/d1a64ea6-aacd-4286-b62d-464763e9333cn%40googlegroups.com.

[gayane.ag...@gmail.com]

Dear Chris,

I'm getting this error after make_brainGraphList on probtrackx data using RStudio V. 1.4.1106, R v. 4.1.2 and brainGraph 3.0.2 on a macOS Monterey 12.1.

Could you help me with this?

make_brainGraphList(A.norm.sub[[1]][,,1], atlas, modality=modality,
+                     weighting='sld', threshold=thresholds[1],
+                     weighted=TRUE, gnames=covars.dti$Study.ID[1],
+                     groups=covars.dti$Group[1])
Error in make_brainGraphList.array(A.norm.sub[[1]][, , 1], atlas, modality = modality,  :
  length(gnames) == kNumGraphs is not TRUE
> traceback()
4: stop(simpleError(msg, call = if (p <- sys.parent(1L)) sys.call(p)))
3: stopifnot(length(gnames) == kNumGraphs)
2: make_brainGraphList.array(A.norm.sub[[1]][, , 1], atlas, modality = modality,
       weighting = "sld", threshold = thresholds[1], weighted = TRUE,
       gnames = covars.dti$Study.ID[1], groups = covars.dti$Group[1])
1: make_brainGraphList(A.norm.sub[[1]][, , 1], atlas, modality = modality,
       weighting = "sld", threshold = thresholds[1], weighted = TRUE,
       gnames = covars.dti$Study.ID[1], groups = covars.dti$Group[1])
> 

Thanks a lot

Gayane

[Chris Watson]

Try changing "groups" in the function call to "grpNames".

Chris

--

You received this message because you are subscribed to the Google Groups "brainGraph-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brainGraph-he...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/6b3a18ab-9b75-4334-88bf-f0b53363c0e2n%40googlegroups.com.

[gayane.ag...@gmail.com]

Dear Chris,

After a long break, I'm back to this analysis again.

I've changed the  "groups" in the function to "grpNames", but now I'm getting another error.

traceback()
5: stop(simpleError(msg, call = expr))
4: e$fun(obj, substitute(ex), parent.frame(), e$data)
3: foreach(i = seq_len(kNumGraphs)) %dopar% {
       res <- loopfun(x[, , i], atlas, type, level, set.attrs, modality, 
           weighting, threshold[i], name = gnames[i], Group = grpNames[i], 
           subnet = subnet, mode = mode, diag = diag, weighted = weighted, 
           use.parallel = FALSE, ...)
   }

2: make_brainGraphList.array(A.norm.sub[[1]], atlas, modality = modality, 

       weighting = "sld", threshold = thresholds[1], weighted = TRUE, 

       gnames = covars.dti$Study.ID, grpNames = covars.dti$Group)
1: make_brainGraphList(A.norm.sub[[1]], atlas, modality = modality, 

       weighting = "sld", threshold = thresholds[1], weighted = TRUE, 

       gnames = covars.dti$Study.ID, grpNames = covars.dti$Group)

You can access my workspace data with this link, if needed: 

https://drive.google.com/file/d/1gYlX77kSov7Mv5gZ3OqlSmQ_Z7sYAiCT/view?usp=sharing

Very much appreciate your help,

Thanks a lot,

Gayane

[Chris Watson]

I apologize for such a long delay; I hadn't been able to devote time to the package. But I will be able to now.

Gayane, what was the exact error message printed to the console (not just the output of "traceback()", which can be quite cryptic)?

Chris

--
You received this message because you are subscribed to the Google Groups "brainGraph-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brainGraph-he...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/brainGraph-help/7770eb97-3384-419c-b8b1-c143a1e10177n%40googlegroups.com.