Fortran runtime error on Ubuntu 16.04

[IdoBar]

Hi,

I'm trying to run Sequoia on an Ubuntu server (16.04, running Microsoft R Open 3.3.2, with the current github version of Sequoia, 0.5). This is my command: ParOUT <- sequoia(GenoFile = GenoDataFile,LifeHistFile = "LifeHistFile_assigned_SA.txt", NumSibRounds=10, MaxSibshipSize=10000, MaxMismatch = 20, WriteFiles = TRUE)

And that's the output:

At line 7488 of file Sequoia.f90
Fortran runtime error: No such file or directory

I do get AgePriors.txt and SequoiaSpecs.txt in sequoiaOUT/ folder, so these functions work.

Same function with the same data works on windows.

Thanks, Ido

[JiscaH]

Hi Ido,

I think the problem is that it should read

sequoia(GenoFile = "GenoDataFile.txt", ....)

Please note that Sequoia is now also on CRAN, so you can download a slightly newer version via the standard install.packages(). That version works fine on Lubuntu 16, but for some reason not reliably on Debian, so I'm not sure if it will work for you. I am working on a fix, and hope to have that on CRAN in a few days.

The CRAN version doesn't rely on reading/writing to files, which so should hopefully avoid problems like you're having. You can find the updated vignette on https://CRAN.R-project.org/package=sequoia https://cran.r-project.org/package=sequoia.

Cheers, Jisca

On 08/03/2017 05:38, Ido Bar wrote:

Hi,

I'm trying to run Sequoia on an Ubuntu server (16.04, running Microsoft R Open 3.3.2, with the current github version of Sequoia, 0.5). This is my command: |ParOUT <- sequoia(GenoFile = GenoDataFile,LifeHistFile = "LifeHistFile_assigned_SA.txt", NumSibRounds=10, MaxSibshipSize=10000, MaxMismatch = 20, WriteFiles = TRUE)|

And that's the output:

|At line 7488 of file Sequoia.f90 Fortran runtime error: No such file or directory |

I do get AgePriors.txt and SequoiaSpecs.txt in sequoiaOUT/ folder, so these functions work.

Same function with the same data works on windows.

Thanks, Ido

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/JiscaH/sequoia/issues/1, or mute the thread https://github.com/notifications/unsubscribe-auth/AQwkHQpID83jsVuPtMAZPTHgQGzJ8dEbks5rjj7igaJpZM4MWYzt.

[IdoBar]

Hi Jisca,

It's not a problem with the file name (the GenoDataFile is just a character string containing the file name and location) and as I said, it's working perfectly under Windows. I'll try the CRAN version and will see if it's working (indeed it was a bit strange having to save my objects to file, rather than just pointing to them in the function).

Thanks, Ido

[JiscaH]

Hello Ido,

A tarball of the new version is now available on CRAN, https://cran.r-project.org/package=sequoia . It may take a few days until they update the windows version.

Please don't hesitate to contact me with any further questions or comments.

Kind regards, Jisca

On 08/03/2017 13:52, Ido Bar wrote:

Hi Jisca,

It's not a problem with the file name (the GenoDataFile is just a character string containing the file name and location) and as I said, it's working perfectly under Windows. I'll try the CRAN version and will see if it's working (indeed it was a bit strange having to save my objects to file, rather than just pointing to them in the function).

Thanks, Ido

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/JiscaH/sequoia/issues/1#issuecomment-285045746, or mute the thread https://github.com/notifications/unsubscribe-auth/AQwkHctrhAbvR4ALGcsKpa5LCdCa2kSnks5rjrK7gaJpZM4MWYzt.

[IdoBar]

Hi Jisca,

I tried the new CRAN version (v0.8.1), using data.frames for GenoData and LifeHistData inputs, however, this time I get a different error:

Error in r[i1] - r[-length(r):-(length(r) - lag + 1L)] : 
  non-numeric argument to binary operator

Seems to me like it should have been r[i,1] in the code, but I can't verify it.

I'll roll back to the github version for now and work on windows.

Thanks, Ido

[JiscaH]

Hi Ido,

I am so sorry about this! The bit of code is not mine but one of the functions I'm calling - I'm just not sure which.

Two likely causes:

• The data is not read into R properly. If you're using the GenotypeData.txt you made earlier, please use GM <- as.matrix(read.table("GenotypeData.txt", row.names=1))

• There is an NA, or other non-number, in either GenoData or LifeHistData. Could you please check if this is the case, and if so, replace the NA's (or otherwise) by any negative number? I've tried out putting NA's in various places in the test datasets but can't replicate the problem, so I'm afraid I can't be more specific.

Does the example code at the bottom of ?sequoia work OK for you?

If you want, I can have a look at your data and try to figure out what the problem is; I promise I would delete the data straight afterwards.

I hope this helps, Jisca

On 14/03/2017 04:40, Ido Bar wrote:

Hi Jisca,

I tried the new CRAN version (v0.8.1), using data.frames for GenoData and LifeHistData inputs, however, this time I get a different error:

|Error in r[i1] - r[-length(r):-(length(r) - lag + 1L)] : non-numeric argument to binary operator |

Seems to me like it should have been |r[i,1]| in the code, but I can't verify it.

I'll roll back to the github version for now and work on windows.

Thanks, Ido

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/JiscaH/sequoia/issues/1#issuecomment-286320428, or mute the thread https://github.com/notifications/unsubscribe-auth/AQwkHbcWsjkOK5ds_jLwUvwOx6cJJx0Iks5rlho3gaJpZM4MWYzt.

[IdoBar]

Hi Jisca,

Thank you for your help, the issue was (as you've suggested), due to character values in LifeHistData (both Year and Sex were read as characters from the file). It is working now using the recent CRAN version.

Thanks, Ido