UnicodeDecodeError about longqc HOT 6 CLOSED

Same problem. Started today with the app. Any help will be appreciated.

from longqc.

this may be caused by special character or specification change in PacBio bam file.
Just curious, your bam file is a file dumped by Sequel? or is there any intermediate step?
will check at my side.

In the meantime, if you convert your bam to fastq/fasta format, problem might disappear.

Yoshinori

from longqc.

Hi Yoshinori,

Yes, the bam file was dumped by sequel "subread.bam". As I know this the file is ready for downstream analyses. I will try samtools conversion to fasta. Thank you, I will let you know the final result.

from longqc.

Hello

I got the bam files from: https://github.com/PacificBiosciences/DevNet/wiki/Microbial-Multiplexing-Data-Set---48-plex:-PacBio-Sequel-II-System,-Chemistry-v2.0,-SMRT-Link-v8.0-Analysis

I changed the file names to make them shorter.

Converting the bam files to fastq worked perfectly.

Thanks for you help!

from longqc.

Thanks both, that sounds some encoding changes happen on PacBio bam file. I will check with PacBio people when I have a chance.
The problematic code just checks if a given bam is gzipped or not, but maybe pacbio bam is not being encoded in standard UTF-8 anymore.
will check the provided file, much appreciated!

Y.

from longqc.

PacBio seems to have started using compressed BAMs rather than uncompressed BAMs as a default. This hits a bug in the code - l_text (uint32_t) was loaded as a string of UTF-8 (my misunderstanding); hence decoding error.
Now only 4 bytes are read from a file and checked.

Y.

from longqc.