novoalab / EpiNano

Detection of RNA modifications from Oxford Nanopore direct RNA sequencing reads (Liu*, Begik* et al., Nature Comm 2019)
GNU General Public License v2.0
109 stars 31 forks source link

Epinano differr warning #129

Closed rania-o closed 1 year ago

rania-o commented 1 year ago

Hello, Each time I use Epinano_DiffErr.R, I get this warning :

1: ggrepel: 23 unlabeled data points (too many overlaps). Consider increasing max.overlaps 
2: ggrepel: 42 unlabeled data points (too many overlaps). Consider increasing max.overlaps

Please, can you tell me what this means ? and how to increase these overlaps ? and if I do, will it enhance the modif detection ?

Rania

enovoa commented 1 year ago

Hi @rania-o - have you tried the test data of the repo? Thanks

rania-o commented 1 year ago

Hi @enovoa Thank you for your fast reply. No I didn't use it with the test data of the repo, but I used with other test data and it worked well without the warning.

enovoa commented 1 year ago

Please share reproducible command line with input data to be able to reproduce the error, because we have never encountered it. Thank you!

Huanle commented 1 year ago

Hi @enovoa Thank you for your fast reply. No I didn't use it with the test data of the repo, but I used with other test data and it worked well without the warning.

Hi @rania-o ,

Have you sorted it out? If you haven't, do you mind sharing your input data?

Best, Huanle

rania-o commented 1 year ago

Hi @Huanle , @enovoa

Sorry for my late reply. No, I didn't sorted it out, but as I already mentioned I do get results with the warning. Our data is too big to transfer it to you. On the other hand, I got the warning also by testing the data provided by Drummer in their github, here is the command I used:

Rscript /home/rania/softs/EpiNano/Epinano_DiffErr.R -k exome.Ad5.UNMOD.plus_strand.per.site.csv -w exome.Ad5.MOD.plus_strand.per.site.csv -o epinano_results_001_deviance_4_t -f sum_err -p -t 4 -d 0.01

You can find the *.csv files of the command Epinano_Variants.py.

Thank you, Rania exome.Ad5.UNMOD.plus_strand.per.site.csv

exome.Ad5.MOD.plus_strand.per.site.csv

Huanle commented 1 year ago

Hi @rania-o ,

It is an alarming emitted by ggrepel, which repels away texts, labels and data points. But when there are too many overlaps between the above mentioned elements, it relseases the warning message you have seen.

It wont affect the analysis. If you are annoyed by the warning message, you can add ggrepel.max.overlaps = Inf (this suppresses the warning msg globally) to the script.

Hope this helps.

Cheers - Huanle