[maker-devel] BLAST options error / should Maker check for file format?

[Carson Holt]

Hi Malte,

Actually there already is.  I’m very surprised your file made it that far.
Normally it fails right away.

Example —>

STATUS: Parsing control files...
STATUS: Processing and indexing input FASTA files...
ERROR: The fasta file /Users/cholt/Developer/maker/trunk/data/test1
appears to be empty.


Another test file —>


STATUS: Parsing control files...
STATUS: Processing and indexing input FASTA files...
ERROR: The nucleotide sequence file
'/Users/cholt/Developer/maker/trunk/data/test2'
appears to contain protein sequence or unrecognized characters. Note
the following nucleotides may be valid but are unsupported [RYKMSWBDHV]
Please check/fix the file before continuing, or set -fix_nucleotides on
the command line to fix this automatically.
Invalid Character: 'M'




You seem to have found just the right formula of improper input to get
past the filters on your run :-)


Thanks,
Carson




On 2/13/14, 8:00 PM, "Malte Petersen" <mptrsen at uni-bonn.de> wrote:

>Dear MAKER devs,
>
>I was running Maker version 2.30p-beta on an insect genome, and it
>didn't produce any output. I got these error messages:
>
>
>Widget::formater:
>/path/to/makeblastdb -dbtype nucl -in
>/tmp/maker_wwA6WO/0/blastprep/120215_I277_FCD0KP1ACXX_L7_INSjdsTAURAAPEI-6
>2_e3%2Escaf.mpi.10.0
>#-------------------------------#
>BLAST options error: File
>/tmp/maker_wwA6WO/0/blastprep/120215_I277_FCD0KP1ACXX_L7_INSjdsTAURAAPEI-6
>2_e3%2Escaf.mpi.10.0
>is empty
>ERROR: /path/to/makeblastdb failed in Widget::formater
>--> rank=NA, hostname=Jeanne-GBR
>ERROR: Failed while doing blastn of ESTs
>ERROR: Chunk failed at level:0, tier_type:3
>FAILED CONTIG:scf7180005143343
>
>ERROR: Chunk failed at level:4, tier_type:0
>FAILED CONTIG:scf7180005143343
>
>
>I figured out that this error is due to a non-Fasta file format being
>fed to Maker as extrinsic evidence (I gave it a meta-info file).  While
>I got the pipeline running now with the correct file, I think that it
>should be complaining (a lot earlier) if any of the input files are of
>the wrong format.  More people might run into this problem and have no
>idea where to look for a solution.
>
>What do you think?
>
>Best,
>Malte
>
>
>_______________________________________________
>maker-devel mailing list
>maker-devel at box290.bluehost.com
>http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org