Could not connect to ENSEMBL about drop HOT 14 CLOSED

Hi, I got the same error when running it last week. It seems there are some problems/updates with the ENSEMBL database. I simply tried the following day and it worked. It could be a persisting problem from this: http://www.ensembl.info/2020/02/18/reduced-functionality-16th-25th-march-2020/

from drop.

Because it crashed, how do I rerun the pipeline? Do I need to rerun from scratch? Or is snakemake smart enough to know that it needs to rerun only this part?

from drop.

snakemake is smart enough to know that it needs to rerun only that and the following scripts. You can check it by doing snakemake -n aberrantSplicing

from drop.

I reran and it completed without error, but the results table is empty. Can you take a look at the results?
https://www.dropbox.com/sh/4mp8jkwnk9vzkkb/AACQ1LHEE7m7pH5vVO_OEbtka?dl=0

from drop.

if you look at plot 3.1 https://www.dropbox.com/sh/4mp8jkwnk9vzkkb/AACsAmCfG2BXKDB0by8M-OLca/AberrantSplicing/GTEX100_summary_files/figure-html?dl=0&preview=unnamed-chunk-3-1.svg&subfolder_nav_tracking=1, it shows a median of 30 aberrantly spliced genes for each splice type (psi 5, psi 3 and theta), so the results table shouldn't be empty.
Did you check both the results_per_junction and results files? Are both of them empty?

from drop.

All I see in the FRASER -> Datasets.R web page is:
"Results
The results table can also be downloaded with the link below.
Download results table
Results table --> EMPTY"

Where can I find results_per_junction and results files?

from drop.

go to /processed_data/aberrant_splicing/results/
there are the .tsv files containing the results

from drop.

GTEX100_results.tsv is empty
GTEX100_results_per_junction.tsv has results.

Is the problem with the ENSEMBL issue? Did that cause GTEX100_results.tsv to be empty?
I reran the pipeline, but maybe snakemake didn't know to rerun that part of the pipeline that failed before, and then that caused GTEX100_results.tsv to be empty?

from drop.

can you check the hgncSymbol column in the results by junction, does it have values?
The ENSEMBL annotation happens in the same script as the extraction of the results. It could be that even though the connection to ENSEMBL worked, it was empty.

from drop.

hgncSymbol is all = 'NA'

from drop.

Hi, Any update about this bug?

from drop.

I was not able to reproduce this error. My guess is that, even though the connection to ensembl worked, it didn't give any results. Rerunning it should fix it.
To do that please remove the following files:
<project_dir>/.drop/tmp/AS/done
<project_dir>/.drop/tmp/AS.done
/*.html # all html files, the directory is specified in the config
/processed_data/aberrant_splicing/results/ # the whole directory

Then rerun the aberrant splicing module. Hopefully the connection to ensembl will work and the genes will get annotated.

from drop.

were you able to update the module and rerun it successfully?

from drop.

Yes this error was fixed.

from drop.