2
0
Entering edit mode
@mariandhore-9275
Last seen 5.9 years ago
United Kingdom

Hello,

I then run into trouble when I attempt to create the summary object. This is the command I used: fast5files_Summary <- readFast5Summary(fast5files)

And I get the following error:

Checking file validity
Reading Channel Data
Reading Raw Data
Reading Template Data
Reading Complement Data
Reading Template FASTQ
Error in strsplit(strings, "\n") : non-character argument

I used the following command to read in the list of files: fast5files <- list.files(path = "Y:/MinION/data", pattern = ".fast5\$", full.names = TRUE)

I'm using version 1.0.0.

Marian

R sequencing preprocessing bioconductor • 1.0k views
0
Entering edit mode
Mike Smith ★ 5.4k
@mike-smith
Last seen 20 hours ago
EMBL Heidelberg / de.NBI

Hi Marian,

Thanks for the interest in the IONiseR package.  I have two possible suggestions at the moment.  My initial feeling is that the problem may be related to the specific files you're feeding to the function - they may not actually be the fast5 files that contain the read data.

In my experience the Nanopore outputs files in the top level folder that are related to the setup of the device prior to sequencing, and may also contain some metrics produced during the run.  These seem to have names that include things like 'mux_scan', although I've not seen a comprehensive documentation of them.  They can also be very large, sometimes several gigabytes.  Unfortunately these all have the file extension .fast5, but internally they don't really look like the typical read files.

The actual read fast5 files tend to end up in a subfolder called 'downloads' and then 'pass' or 'fail' - you may be interested in both of these.  A good run typically produces several thousand fast5 files, but if things didn't work very well it can be as low as single digits.  Your Metrichor account will tell you how many reads were produced, an whether they ended up in the pass or fail category.

The other alternative is that your files haven't been to the base caller, and IONiseR is falling over when it tried to extract a non-existent FASTQ field.  In which case I need to provide a more helpful error message!

If you can share the files on an FTP or something I am happy to take a look and see if I can shed some more light on it.

Mike

0
Entering edit mode
Mike Smith ★ 5.4k
@mike-smith
Last seen 20 hours ago
EMBL Heidelberg / de.NBI

Hi Marian,

Thank you for sending your files to me.  It looks like Oxford Nanopore have updated the structure of the fast5 files, and IONiseR was no longer finding things in the places it expected.  I have updated the code in the devel release of Bioconductor, so IONiseR verson 1.1.1 can now read the files you sent me.  It should also print some more informative error messages if this happens again.

Many thanks for providing the data, I couldn't have done this without your help.  Let me know if you have any more queries.

Thanks,

Mike

0
Entering edit mode

Hi Mike,

Thank you very much.  I'll let you know how I get on with the new version.

Many thanks,

Marian

0
Entering edit mode

Thanks Marian.  Just to let you know, since this was a show stopping problem I've also patched the release version of IONiseR, so you can use version 1.0.1 if you prefer not to work with developmental packages.