Closed sarah872 closed 4 years ago
Hi Sarah,
I think the issue comes from the fact the your contain files are named differently from the corresponding annotation folders :
spades_contigs/annotation/P20_strain
while your contig file names are of the kind:
spades_contigs/contigs/P20_strain.contigs.fasta
The pipeline script tries to match these; doing so, it will trim the .fasta
extension from the contig file name, but will leave in the .contigs
string.
So trying to match P20_strain.contigs
from the contig file with an annotation folder, it will fail because the folder is only named P20_strain
.
if you check the input data section of the README, you'll see an example.
Sorry I know the formatting is a bit tedious, but you almost got it right! I though to introduce non-exact string matching to ease this sort of cases, but thought better not because bacterial strain names are so random and often similar to each others, there could be a chance match with the wrong assembly.
I hope this helps.
best wishes,
Florent
Thanks a lot! Unfortunately, there are still some errors.
found 10 contig files (raw genome assemblies) in spades_contigs/contigs/
[2019-09-19 11:44:57] I15_strain.contigs
found annotation folder 'spades_contigs/annotation/I15_strain.contigs' ; skip annotation of contigs in 'spades_contigs/contigs/I15_strain.contigs.fasta'
ln: failed to create symbolic link ‘spades_contigs_db/spades_contigs_db/00.input_data/annotation/’: File exists
fix annotation to integrate region information into GFF files
ls: cannot access spades_contigs_db/spades_contigs_db/00.input_data/annotation/I15_strain.contigs/*.gff: No such file or directory
Error: missing mandatory GFF file in spades_contigs_db/spades_contigs_db/00.input_data/annotation/I15_strain.contigs/ folder
ERROR: Pantagrel pipeline task 0: failed.
I deleted the database folder spades_contigs_db
and re-run the database generation, but without any effect.
so this was a bug! it should be fixed with 2e98c27. you can update the git repository and rerun the task ; if you choose not to erase the whole database folder, please do not forget to run first:
pantagruel -i spades_contigs_db/spades_contigs_db/environ_pantagruel_spades_contigs_db.sh --refresh init
Still an error:
ls: cannot access db_name/00.input_data/annotation/I15_strain.contigs/*.gff: No such file or directory
it seems that the symbolic link is created wrong:
lrwxrwxrwx. 1 user login 23 Sep 19 15:42 db_name/00.input_data/annotation/I15_strain.contigs -> /I15_strain.contigs
the I15_strain.contigs
is not in the root dir.
loops.... sorry the fix was dirty there was a typo in the line ; corrected here 23eab0b.
I want to run
pantagruel
on 10 genomes. This is the directory structure of my input dir:I am running
pantagruel
like this:The pipeline is starting, but after running prokka, there is an error: