search

BoevaLab / FREEC

Control-FREEC: Copy number and genotype annotation in whole genome and whole exome sequencing data
153 stars 49 forks source link

mateOrientation or matesOrientation? #89

Open rsharris opened 3 years ago

rsharris commented 3 years ago

The program is giving me this error:

Error: FREEC was not able to extract reads from ./pacbio.pileup.gz
Check your parameters: inputFormat and mateOrientation
Use "matesOrientation=0" if you have single end reads

Notice that it is advising me to check mateOrientation (no "s") but to use matesOrientation (with an "s").

My first thought was that matesOrientation is a typo. After all, mateOrientation appears in the manual at http://boevalab.inf.ethz.ch/FREEC/tutorial.html while matesOrientation does not.

However, I do see in the code that both mateOrientation and matesOrientation are read from the config file.

valeu commented 3 years ago

yes, both are read because of the initial typo.

Try to provide the full path to the pileup.gz file, please?

rsharris commented 3 years ago

Error: FREEC was not able to extract reads from ./pacbio.pileup.gz

My best guess at the moment is that freec was able to read the file but found no reads in it. This might have been the result of using the --no-BAQ when I ran mpileup, which I did to avoid runtime and memory issues mpileup apparently has when dealing with long reads. (without --no-BAQ, mpileup ran for 4 days then failed because it was out of memory).

So I am currently investigating that.

yes, both are read because of the initial typo.

I think it would be an improvement to change that print statement to say Use mateOrientation=0....

As it stands, it's telling the user to use an option that isn't in the documentation. When it told me that I naturally wanted to find out what that option did. My config file was already using mateOrientation=0. So instead of trying to figure out what was wrong with my pileup file, I had to spend time trying to figure out what "matesOrientation" does.

Try to provide the full path to the pileup.gz file, please?

I simplified the path in my post. The actual path I have in my config file is ../mapped_reads/pacbio.pileup.gz.

In earlier attempts I had, in fact, given it the full path. But for some reason it wasn't able to find the file at all -- I got an error like "unable to open file /pacbio.pileup.gz". Because I am running on a slurm cluster, there are some differences in the file hierarchy between the cluster nodes and the host machine from which I submit the job. (The path I had given was valid on the host machine.) Instead of trying to resolve that, I decided to give freec a relative path to the file.