guyleonard/orchard

:seedling: :clipboard: :arrow_right: :8ball: :arrow_right: :palm_tree: :evergreen_tree: :deciduous_tree: - A quick phylogenetic tree building pipeline!

orchard's Issues

trimal related

Using trimal 1.4 has given this error:
Error: the symbol 'K' accesing the matrix is not defined in this object
but I do not know why! The data looked fine.
Trimal v1.2 works fine...

special taxa doesn't work

special taxa won't run if there is only one taxa - split doesn't split

gzip report folder

Might as well gzip the reports in that folder as they are rarely viewed and take up space, gzip or bzip them!

filter out final trees if they don't contain X or Y taxa

Used to have an option to ignore trees unless they contained taxa from a NCBI taxon_id, -f option with text file.

blast legacy special taxa

blast legacy is missing the support to blast with high/low-er returns for special taxa!

Selenocysteine replaces U with X in accessions

This is not the behaviour I need from this regex, accessions should not be changed as then they cannot be renamed...

remove file from process if zero sized

zero sized files make the process fail downstream - you can't trim no data - but it throws an exception and exits rather than continuing to the next. zero sized files come up every now and again due to errors previously... so it's best to error report them and move on rather than stopping...

multi-seq/orthogroup puts blast reports as single files rather than in original grouping

something to think about, will be nicer for output but not exactly priority

run_id is confusing

Files are output in the directory "run_id" from the YAML file, perhaps this should be called "output_dir" or something to be more clear....

Moreover, we could keep a run ID and have an output dir as well... that way you can run multiple runs of the same data in one project directory: e.g.
your_project/
your_project/1e-20
your_project/1e-30

read in ortho groups

Used to have an option to read in multiple files associated with orthogroups and output one tree... re-add?

Quit during trimal - "composed only by gaps" error

WARNING: Removing sequence '576696987' composed only by gaps

WARNING: Output alignment has not been generated. It is empty.

------------- EXCEPTION: Bio::Root::Exception -------------
MSG: Can not open '/storage/home/aurelie/hypho_orthoMCL_tree/hypho_node_60_66_orthologues/masks/hyc3ortho122514_hits.afa-tr' for reading: No such file or directory
STACK: Error::throw
STACK: Bio::Root::Root::throw /usr/share/perl5/Bio/Root/Root.pm:486
STACK: Bio::Tools::GuessSeqFormat::guess /usr/share/perl5/Bio/Tools/GuessSeqFormat.pm:464
STACK: Bio::SeqIO::new /usr/share/perl5/Bio/SeqIO.pm:395
STACK: main::mask_check orchard.pl:368
STACK: main::run_trimal orchard.pl:356
STACK: main::masking_step orchard.pl:309

STACK: orchard.pl:181

This appears if the "-nogaps" option runs and does not produce a file... perhaps it would just be easier to go with "-automated1" all the time...

check for prerequisites

script does not check for prerequisites and does not gracefully exit if they don't exist

orchard.pl line 531

Well this should not be a hard link to a local file - read the taxonomy information from the YAML!!

selenocysteines are not being replaced

I have a sed command that is supposed to check for selenocysteine (U) and other junk in the sequence files of each run - as MAFFT has a hissy fit when it sees them and the "--anysymbol" seems to produce shit alignments. The consequence is zero-sized files and then downstream programs quit/crash and then the pipeline fails. needs to be sorted! 1) make sure the U and non-IUPAC protein codes are repalced by X and 2) fail gracefully!!!!!

include other search sequences from seeds...?