Closed ZhenwenYu closed 2 years ago
Hi,
I have quickly pushed a new sub-release (GeneSCF-v1.1-p3). Please use this new version. It will work. Let me know if you still have the issue.
Thanks a lot for the quick response!
Now gene mapping works well, and the output list file is no longer empty. However, the output tsv file is still empty. Is it due to the illegal division by zero at class/lib/List/Vectorize/lib/List.pl line 599,
Please run as "./geneSCF" instead of "perl geneSCF" and see if it works!!
Run as "./geneSCF" still has the same illegal division by zero issues in line 599. I have checked the previous issues closed in the discussion, and found the genes mapped in H0.list, are exactly same as in the KEGG database. My Ubuntu system is freshly installed and is Ubuntu 22.04 LTS, and I don't know if other systems are facing the same issue. Maybe my system version is too new or is still lacking some essential environments?
Can you try giving full paths to the input file and the output directory?
input directory: /home/zhenwen/GeneSCF-master/test output directory: /home/zhenwen/GeneSCF-master/test/output
input directory: /home/zhenwen/GeneSCF-master/test output directory: /home/zhenwen/GeneSCF-master/test/output
I meant try giving the full path to geneSCF run.
Run as "./geneSCF" still has the same illegal division by zero issues in line 599. I have checked the previous issues closed in the discussion, and found the genes mapped in H0.list, are exactly same as in the KEGG database. My Ubuntu system is freshly installed and is Ubuntu 22.04 LTS, and I don't know if other systems are facing the same issue. Maybe my system version is too new or is still lacking some essential environments?
https://github.com/genescf/GeneSCF/issues/4#issuecomment-437520733
Inputting the full path doesn't work either. I am using the Ubuntu Linux system and the default gedit text editor, so reasons 1 and 4 don't make sense. Also, we can see from the terminal that many genes matched the KEGG database, so it couldn't be reason 2. Therefore, could there still be some issues in the' prepare_database' module to update the KEGG database? Moreover, Is there someplace we can use the "try" command to skip the illegal division error message and continue?
Inputting the full path doesn't work either. I am using the Ubuntu Linux system and the default gedit text editor, so reasons 1 and 4 don't make sense. Also, we can see from the terminal that many genes matched the KEGG database, so it couldn't be reason 2. Therefore, could there still be some issues in the' prepare_database' module to update the KEGG database? Moreover, Is there someplace we can use the "try" command to skip the illegal division error message and continue?
Hi,
1) First try giving output directly as /home/zhenwen/GeneSCF-master/test/output/
NOT /home/zhenwen/GeneSCF-master/test/output
. There is some weird thing that some time it doesn't accept without "/" in the end.
2) Second, there is possibility that the KEGG database is not updated properly. Can you try prepare_database
and then try running geneSCF
.
Well, neither of the two options works, but thanks for the help. The bug is still the same, having illegal division by zero issues in line 599. I am emailing a lab whose paper used GeneSCF and is what I want to repeat. I may need to wait for a couple of days to see if they can also prvide some advice.
Hi,
Here is the working geneSCF. There was an issue connecting with KEGG. Now it works fine. Let me know. https://github.com/genescf/GeneSCF/archive/refs/tags/v1.1-p3.beta.tar.gz
You can also download this from latest release.
Thanks a lot for all the helps! Now everything works well with KEGG.
Hi, and thanks for creating GeneSCF,
We are using GeneSCF to annotate genes with the KEGG database. However, we occurred an issue that 0 genes can be mapped, as shown in the attached screenshot. We guess there is an issue in retrieving the KEGG database, as the prepare_database retrieves 0 pathways and the kegg_database.txt in class/lib/db/hsa folder is empty. Could you please tell us how to make it works?
Best, Zhenwen