length of snp vector should be 1 less than ncol(res$lbf_variable)

Question

length of snp vector should be 1 less than ncol(res$lbf_variable)

shannjiang opened this issue a year ago · 6 comments

Hi,

I bumped to this error when running susie on the attached test data:

** Examples

library(coloc)
data(coloc_test_data)
result=runsusie(coloc_test_data$D1)
running max iterations: 100
converged: TRUE
Error in annotate_susie(res, snp): length of snp vector should be 1 less than ncol(res$lbf_variable)
summary(result)
Error in summary(result): object 'result' not found

What is the ncol(res$lbf_variable) which should be greater than the length of snp vector specifically?

Thanks,
Shan

Answer 1 · 2023-08-22T10:48:44.000Z

can you please download the latest version from github? This issue should be fixed.

Answer 2 · 2023-08-22T17:17:21.000Z

Hi @chr1swallace ,

Thanks for figuring that out.
After I successfully run the command runsusie(), the command summary(result) didn't show anything informative:

result=runsusie(coloc_test_data$D1)
running max iterations: 100
converged: TRUE
> summary(result)
Mode FALSE
logical 1

And also if I run the runsusie() on the D3 in the test data I got the error 'LD' object not found:

result=runsusie(coloc_test_data$D3)
running max iterations: 100
converged: TRUE
Error in .susie_setld(res$sets$cs, LD) : object 'LD' not found

But I checked the structure of D3, LD is there:

str(coloc_test_data$D3)
List of 9
$ beta : num [1:500] 0.343 0.502 0.541 0.233 0.536 ...
$ varbeta : num [1:500] 0.02321 0.0073 0.00991 0.0205 0.00875 ...
$ N : int 1000
$ sdY : num 1.31
$ type : chr "quant"
$ MAF : num [1:500] 0.0405 0.149 0.0955 0.0485 0.11 0.073 0.0795 0.038 0.234 0.131 ...
$ LD : num [1:500, 1:500] 1 0.456 0.611 0.906 0.559 ...
..- attr(*, "dimnames")=List of 2
.. ..$ : chr [1:500] "s1" "s2" "s3" "s4" ...
.. ..$ : chr [1:500] "s1" "s2" "s3" "s4" ...
$ snp : chr [1:500] "s1" "s2" "s3" "s4" ...
$ position: int [1:500] 1 2 3 4 5 6 7 8 9 10 ...

And I tested on my own data which also prompted me 'LD' object not found but I am sure it's in my data.

Thanks,

Shan

Answer 3 · 2023-08-22T17:33:14.000Z

It's not your data, sorry. I moved some code to help a user with a specific request and silently broke something. Now fixed. Could I ask you to download afresh from GitHub?

-- https://chr1swallace.github.io

________________________________ From: shannjiang ***@***.***> Sent: Tuesday, August 22, 2023 6:17:31 PM To: chr1swallace/coloc ***@***.***> Cc: Chris Wallace ***@***.***>; Mention ***@***.***> Subject: Re: [chr1swallace/coloc] length of snp vector should be 1 less than ncol(res$lbf_variable) (Issue #127) Hi @chr1swallace<https://github.com/chr1swallace> , Thanks for figuring that out. After I successfully run the command runsusie(), the command summary(result) didn't show anything informative: result=runsusie(coloc_test_data$D1) running max iterations: 100 converged: TRUE

summary(result)

Mode FALSE logical 1 And also if I run the runsusie() on the D3 in the test data I got the error 'LD' object not found: result=runsusie(coloc_test_data$D3) running max iterations: 100 converged: TRUE Error in .susie_setld(res$sets$cs, LD) : object 'LD' not found But I checked the structure of D3, LD is there: str(coloc_test_data$D3) List of 9 $ beta : num [1:500] 0.343 0.502 0.541 0.233 0.536 ... $ varbeta : num [1:500] 0.02321 0.0073 0.00991 0.0205 0.00875 ... $ N : int 1000 $ sdY : num 1.31 $ type : chr "quant" $ MAF : num [1:500] 0.0405 0.149 0.0955 0.0485 0.11 0.073 0.0795 0.038 0.234 0.131 ... $ LD : num [1:500, 1:500] 1 0.456 0.611 0.906 0.559 ... ..- attr(*, "dimnames")=List of 2 .. ..$ : chr [1:500] "s1" "s2" "s3" "s4" ... .. ..$ : chr [1:500] "s1" "s2" "s3" "s4" ... $ snp : chr [1:500] "s1" "s2" "s3" "s4" ... $ position: int [1:500] 1 2 3 4 5 6 7 8 9 10 ... And I tested on my own data which also prompted me 'LD' object not found but I am sure it's in my data. Thanks, Shan — Reply to this email directly, view it on GitHub<#127 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAQWR2AO2GLEWRF3PRVWBC3XWTSSXANCNFSM6AAAAAA3ACDFVE>. You are receiving this because you were mentioned.Message ID: ***@***.***>

Answer 4 · 2023-08-22T18:02:24.000Z

@chr1swallace the aboved mentioned information is what I updated coloc:

library(coloc)
This is coloc version 5.2.3
data(coloc_test_data)
result=runsusie(coloc_test_data$D1)
running max iterations: 100
converged: TRUE
> summary(result)
Mode FALSE
logical 1
result=runsusie(coloc_test_data$D3)
running max iterations: 100
converged: TRUE
Error in .susie_setld(res$sets$cs, LD) : object 'LD' not found

Answer 5 · 2023-08-22T19:23:27.000Z

so sorry, I failed to push the change to github. try again?

-- https://chr1swallace.github.io<https://chr1swallace.github.io/>

________________________________ From: shannjiang ***@***.***> Sent: Tuesday, August 22, 2023 7:02 PM To: chr1swallace/coloc ***@***.***> Cc: Chris Wallace ***@***.***>; Mention ***@***.***> Subject: Re: [chr1swallace/coloc] length of snp vector should be 1 less than ncol(res$lbf_variable) (Issue #127) @chr1swallace<https://github.com/chr1swallace> the aboved mentioned information is what I updated coloc: library(coloc) This is coloc version 5.2.3 data(coloc_test_data) result=runsusie(coloc_test_data$D1) running max iterations: 100 converged: TRUE

summary(result)

Mode FALSE logical 1 result=runsusie(coloc_test_data$D3) running max iterations: 100 converged: TRUE Error in .susie_setld(res$sets$cs, LD) : object 'LD' not found — Reply to this email directly, view it on GitHub<#127 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAQWR2HC3JEMPYO5YZYLMJDXWTX3VANCNFSM6AAAAAA3ACDFVE>. You are receiving this because you were mentioned.Message ID: ***@***.***>

Answer 6 · 2023-08-28T15:15:23.000Z

Hi @chr1swallace,

Thank you so much! It's working now.

Best,
Shan