Problems installing Bio++ and MafFilter

[Marcos Barbeitos]

Hello,

I am trying to use the installation script in <http://biopp.univ-montp2.fr/Download/bpp-setup.sh to install Bio++> in my local machine (Linux Mint 17.1) and I am getting the following message when installing bppsuite-2.2.0:

-- Found nroff here: /usr/bin/nroff

   Adding targets: man

CMake Error at CMakeLists.txt:134 (MESSAGE):

  p"Program makeinfo required but not found."

Also, when trying to install MafFilter using cmake I get:

  Could not find a package configuration file provided by "bpp-phyl-omics"

  (requested version 2.0.0) with any of the following names:

    bpp-phyl-omicsConfig.cmake

    bpp-phyl-omics-config.cmake

  Add the installation prefix of "bpp-phyl-omics" to CMAKE_PREFIX_PATH or set

  "bpp-phyl-omics_DIR" to a directory containing one of the above files.  If

  "bpp-phyl-omics" provides a separate development package or SDK, be sure it

  has been installed.

I've searched my computer for both files and they are nowhere.

I found a bug report on this issue in <https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=865901> but it just states that the bug should be fixed in newest version at the Debian FTP archive.

Should I just try to pull the latest version out of GitHub and compile from source?

Cheers!

[Julien Yann Dutheil]

Dear Marcos,

As we moved to GitHub, the bppSetup.sh script is kind of deprecated, or at least has not been updated to version 2.3.0. I recommend that you install version 2.3 from GitHub. There are also static executable for maffilter available https://github.com/jydu/maffilter/releases , which do not require Bio++ to be installed, if that would help. Also note that debian packages are available for bio++ 2.3 and maffilter (from Debian Testing), they should probably work on Linux Mint.

Best regards,

Julien.

[Marcos Barbeitos]

Dear Julien,

Thank you so much for your reply, the binary indeed works in my machine.

I intend to use MafFilter to eliminate duplicates from MAF files generated by pairwise genome alignments. 

Several of these genomes are from non-model organisms, so many of them do not have chromosome information.

I wonder what MafFilter expects as the sequence header.

For instance, the Homo sapiens assembly we are working with has headers such as:

>NC_000001.11 Homo sapiens chromosome 1, GRCh38.p7 Primary Assembly

but in the MAF file, only the accession number was retained (i.e. NC_000001.11). 

Because MafFilter requires the species name, how can we modify the sequence names in the MAF files so that they meet what the parser expects?

Cheers!

[Julien Yann Dutheil]

Dear Marcos,

The Maf format supposes that sequence names are in the form 'species.chr'. MafFilter therefore expects sthg like this. Yet in practice, chromosome can be contig or scafold, it does not impact the program in any way. You should therefore have sthg like Hsapiens.NC_000001.11.  Note that dots whithin "chromosome" names might also potentially be confusing, I don't remember whether the program will consider the first or last dot as a species separator. In your case however, as you have no species name AND a dot in the contig name, it will assume that species is NC_000001 and chromosome is 11. It is usually a good idea to correct the names before generating the alignment, but I guess you could also do that a posteriori with a series of grep commands or so.

Best,

Julien.