Question

IONiseR error when trying to call readFast5Summary on fast5 reads

0

Entering edit mode

artemd • 0

@artemd-12458

Last seen 7.2 years ago

I saw the last post about this problem (A: IONiseR readFast5Summary Error) and I suspect this problem has returned.

When I run readFast5Summary on a few reads I get this output:

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

I also archived few test reads (only 4 reads) that will throw this error when trying to run readFast5Summary on them. Here is the link https://drive.google.com/open?id=0B8nycJVq5xKReGppdDBHQXpBNjQ

IONiseR • 1.0k views

ADD COMMENT • link updated 7.2 years ago by Mike Smith ★ 6.5k • written 7.2 years ago by artemd • 0

0

Entering edit mode

Is your data recent? There has been an update of the format.

ADD REPLY • link 7.2 years ago WouterDeCoster ▴ 130

0

Entering edit mode

Wouter is correct that this is almost certainly to do with changes in the fast5 file structure. Lots of recent data doesn't follow the 2D calling paradigm, and so there isn't a complement FASTQ entry, and IONiseR falls over. This is particularly true if the files are from the Mux Scan, when you wouldn't expect to generate 2D data (or any real data at all).

I thought I'd patched this error, but it may only be in the developmental branch of Bioconductor. I'll do some more testing this evening and let you know my findings.

ADD REPLY • link 7.2 years ago Mike Smith ★ 6.5k

score 0 · Answer 1 · 2017-02-27

I've committed a patch to IONiseR version 1.4.4 in release (and 1.5.10 in devel) which should cope with this. Thanks very much for providing the example files. Let me know if you encounter any more bugs, I've done some testing, but it's possible some additional functions won't like an object with no complement strand data either.

I'm doing a major rewrite of the package at the moment to try and improve reading performance, and to try and cope more effectively with the the large numbers of slightly different file formats that have been produced over time.