Closed Fnyasimi closed 3 years ago
This is a bug due to version mismatches between gimmemotifs and genomepy. Actually, a new version of gimmemotifs was released today which hopefully also fixes this bug. Can you try this:
conda create -n gimme gimmemotifs=0.14.4
This should create an environment with the correct version.
Regarding your other question: you're correct that this is the equivalent command. It should create a file called gimme.roc.report.txt
in your output directory (when it runs correctly). The reason for the change is two-fold. 1) I wanted a more streamlined experience where the typical analysis could be run with one command. 2) The fact that a genome is required is because it is used to set motif-specific thresholds for scanning. In practice this results in better thresholds than the previous approach (which set too lenient thresholds for short motifs and too strict thresholds for long motifs).
I probably should allow an analysis without a genome , however, this necessitates some code changes which I haven't found the time for.
Thank you the module below is depreciated in sklearn 0.23.1. Thought it will be useful for future releases.
Traceback (most recent call last):
File "/home/festo/miniconda3/bin/gimme", line 8, in <module>
from gimmemotifs.cli import cli
File "/home/festo/miniconda3/lib/python3.7/site-packages/gimmemotifs/cli.py", line 11, in <module>
from gimmemotifs import commands, __version__
File "/home/festo/miniconda3/lib/python3.7/site-packages/gimmemotifs/commands/__init__.py", line 15, in <module>
"{0}.{1}".format(__name__, cmdname), globals(), locals(), [cmdname], level
File "/home/festo/miniconda3/lib/python3.7/site-packages/gimmemotifs/commands/maelstrom.py", line 9, in <module>
from gimmemotifs.maelstrom import run_maelstrom
File "/home/festo/miniconda3/lib/python3.7/site-packages/gimmemotifs/maelstrom.py", line 37, in <module>
from gimmemotifs.moap import moap, Moap, scan_to_table
File "/home/festo/miniconda3/lib/python3.7/site-packages/gimmemotifs/moap.py", line 41, in <module>
from lightning.classification import CDClassifier
File "/home/festo/miniconda3/lib/python3.7/site-packages/lightning/classification.py", line 1, in <module>
from .impl.adagrad import AdaGradClassifier
File "/home/festo/miniconda3/lib/python3.7/site-packages/lightning/impl/adagrad.py", line 8, in <module>
from sklearn.externals.six.moves import xrange
ModuleNotFoundError: No module named 'sklearn.externals.six'
Thanks for the notification. I'm indeed removing this dependency in the current development version. However, this is why I specified scikit-learn=0.22.2.post1
for the current gimmemotifs bioconda package. Not sure how sklearn 0.23 can be installed :/.
The sklearn 0.23.1 can be installed from the conda-forge channel
conda install scikit-learn==0.23.1 -c conda-forge
I was wondering if the meme version used by the tool can be upgraded to v5.1.1?
Sorry for the delay in answering! The new build (gimmemotifs-0.14.4 py37h516909a_1
) should install meme v5.1.1.
Can I ask how you installed / upgraded GimmeMotifs? It seems that some of the depencies are not correctly installed or upgraded. Sometimes with conda the best way to this is to create a fresh new environment:
conda create -n gimme gimmemotifs=0.14.4
I used that command and it comes wit meme v5.0.5
Dependencies should be fixed with the latest release.
Describe the bug I get this error when I try to do Motif enrichment statistics. I downloaded the dm6 genome using genomepy. I dont understand what is causing the problem.
A quick question regarding this can I get the statistics without using the genome? I am interested in replicating the gimme roc command which was depreciated. Previously it was done like this;
gimme roc lola.pfm ENCFF742LVH.fa ENCFF742LVH.bg > lola_gimme_metrics.txt
What is the equivalent of the same?To Reproduce My current try which ran into the error is;
gimme motifs ENCFF742LVH.fa lola-motif -b ENCFF742LVH.bg -p lola.pfm --known -g dm6
Expected behavior I expected to get the
gimme.roc.report.txt
in the results dirError logs
Installation information (please complete the following information):