[BioC] Rbowtie - compilation error

Michael Stadler michael.stadler at fmi.ch
Fri Jun 6 17:30:35 CEST 2014


Hi Romain,

Great, thank you very much for testing.

I am copying the list again so others who might run into the problem are
aware that an update to Rbowtie version 1.4.5 (release) or 1.5.5
(development) should solve the issue.

Best wishes,
Michael


On 05.06.2014 18:06, Romain Retureau wrote:
> Hey,
> 
> I installed the attached version of Rbowtie and it works ! No error, no
> warnings and works on R. 
> 
> Thanks you very much for the fast help ! 
> Romain. 
> 
> 
> 2014-06-05 16:01 GMT+02:00 Michael Stadler <michael.stadler at fmi.ch
> <mailto:michael.stadler at fmi.ch>>:
> 
>     Hi Romain,
> 
>     I think we are getting there. I am replying off-list so I can send you
>     an attachement with a modified version of Rbowtie for you to test.
> 
>     It includes many modifications by the bowtie authors that should fix
>     your problem, however, I have not yet included the POPCNT_CAPABILITY=0.
>     In case this does not do it yet, I will add the POPCNT_CAPABILITY=0 and
>     send you one more version.
> 
>     You can install the attached version of Rbowtie from the command
>     line using:
>       R CMD INSTALL Rbowtie_1.5.5.tar.gz
> 
>     Thanks again for reporting and for testing the modified versions!
>     Best,
>     Michael
> 
> 
> 
>     On 05.06.2014 14:30, Romain Retureau wrote:
>     > Yes, i'm sorry. The "make" didn't works, but the "POPCNT_CAPABILITY=0
>     > make" works.
>     >
>     > After installing it by this command "POPCNT_CAPABILITY=0 make" I did
>     > this "./bowtie --version" and got the message I pasted.
>     >
>     > Sorry, it wasn't clear at all.
>     > Thanks.
>     >
>     >
>     > 2014-06-05 14:08 GMT+02:00 Michael Stadler <michael.stadler at fmi.ch
>     <mailto:michael.stadler at fmi.ch>
>     > <mailto:michael.stadler at fmi.ch <mailto:michael.stadler at fmi.ch>>>:
>     >
>     >     Hi Romain,
>     >
>     >     Please find my replies below:
>     >
>     >     On 05.06.2014 11:05, Romain Retureau wrote:
>     >     > Hi,
>     >     >
>     >     > here is the output of uname -a :
>     >     >
>     >     > /"Linux retureau-MacBookPro 3.2.0-63-generic-pae #95-Ubuntu SMP
>     >     Thu May
>     >     > 15 23:26:11 UTC 2014 i686 i686 i386 GNU/Linux//"/
>     >     > /
>     >     > /
>     >     > I didn't specify it, but i'm running ubuntu in dual bot on a mac
>     >     book pro.
>     >     >
>     >     Fine.
>     >
>     >     > Here is the output of "make" inside the right folder :
>     >     > http://pastebin.com/NZGX9YH0  (it's huge, that's why i'm using
>     >     pastebin).
>     >
>     >     > So I tried with this : "./bowtie --version" and get that :
>     >     >
>     >     > /"./bowtie version 1.0.1/
>     >     > /32-bit/
>     >     > /Built on retureau-MacBookPro/
>     >     > /jeudi 5 juin 2014, 10:57:41 (UTC+0200)/
>     >     > /Compiler: gcc version 4.8.1 (Ubuntu 4.8.1-2ubuntu1~12.04) /
>     >     > /Options: -O3 -m32  -Wl,--hash-style=both  /
>     >     > /Sizeof {int, long, long long, void*, size_t, off_t}: {4, 4,
>     8, 4,
>     >     4, 8}"/
>     >
>     >     Here I am confused. The 'make' output that you pasted shows
>     that bowtie
>     >     is also failing to build, with the same error message as Rbowtie.
>     >
>     >     So it not possible to run './bowtie --version', because there
>     is no
>     >     bowtie binary...
>     >
>     >     In the pasted output, I can find "... -DPOPCNT_CAPABILITY
>     ...", while
>     >     the version below was compiled without that. Am I guessing
>     correctly
>     >     that you compiled with:
>     >       POPCNT_CAPABILITY=0 make
>     >
>     >     Please let me know exactly what you do and where the output
>     comes from,
>     >     otherwise it is difficult for me to help.
>     >
>     >     Michael
>     >
>     >
> 
>     --
>     --------------------------------------------
>     Michael Stadler, PhD
>     Head of Computational Biology
>     Friedrich Miescher Institute
>     Basel (Switzerland)
>     Phone : +41 61 697 6492 <tel:%2B41%2061%20697%206492>
>     Fax   : +41 61 697 3976 <tel:%2B41%2061%20697%203976>
>     Mail  : michael.stadler at fmi.ch <mailto:michael.stadler at fmi.ch>
>     --------------------------------------------
> 
>



More information about the Bioconductor mailing list