opengene / repaq Goto Github PK

repaq使用的俩示例文件是啥？DNA/RNA？人/其他物种？NA12878？想拿这个数据尝试一些好玩的东西，希望了解这个数据的更多信息

% repaq --version
0.2.0

and return 0.

Hello,

Recently, I used repaq to compress a fastq file and then tried to check the result. When I used this cmd: "repaq -p -i xxxxxxx.fq.gz -r test.rfq -j test.txt", it returned "Segmentation fault (core dumped)" and stopped. Could anyone help me to fix it? Thx~

big endian and small endian

I asked for --help so it should return 0 (success) not 1.

Does it need install -d -t PREFIX/bin src/repaq ?

I get the following error message immediately:

xz: (stdin): Cannot allocate memory
ERROR: failed to call xz, please confirm that xz is installed in your system

when I run:

sbatch -c 16 --mem 10000 --wrap "repaq -c -i file_1_trimmed.fastq.gz -I file_2_trimmed.fastq.gz -o file.fpq.xz --compression 9 --thread 16"

I installed repaq and xz with the following commands in a fresh conda environment:

conda install -c bioconda repaq
conda install conda-forge::xz

Help would be much appreciated

Maybe change to install: $(TARGET) so that make install does build and install?

Hi! We're considering using repaq due to the significant space savings, but have some concerns about lossless-ness. Some of the other issues (#11, #12, #13) seem to suggest that repaq isn't lossless, but there isn't enough detail on some (#12, #13) to determine the severity.

How production ready is repaq?

Hello, I recently found some problems using repaq-0.3.0.
After decompressing, I found that the md5 check failed. Comparing the decompressed fastq file with the original file, it was found that at some point in some reads N became G after being compressed. Is this a machine problem or an algorithm problem? I compressed 160 files and 20 of them were problematic. Recompressing these 20 is still the same result.
Here is the result of the --compare parameter
"result":"failed", "msg":"The RFQ file and FASTQ file have different sequence in the 7815 pair. GGACCTCTTCTGACTGATGGGAAATCACAGCAGTTGGAGACCCAGGTCCACAGGAAGGATGAAGAACCCAAGGAATGGCAGCAGACTGAGAGCTTCTGGA | GGACCTCTTCTGACTGATGGGAAATCACAGCAGTTGGAGACCCAGGTCCACAGGAAGGATGAAGAACCCAAGNAATGGCAGCAGACTGAGAGCTTCTGGA"

On some fq files, the quality string is not preserved
in a round-trip conversion. In particular ; or : gets replaced by I.

One example line, first line is original:

-BBDADFHHHHHHII?HHHIIII?EHEHCDHDHGFHHHEFHHHIECD/E?@FH1<@GEGH?GHH?@CC@?E1<D@FGH111G?CHF0CCEG?0<FFE0:F0=
+BBDADFHHHHHHII?HHHIIII?EHEHCDHDHGFHHHEFHHHIECD/E?@FH1<@GEGH?GHH?@CC@?E1<D@FGH111G?CHF0CCEG?0<FFE0IF0=

repaq-command

repaq -c -i original.fq -o original.fq.rfq

since new version of bcl2fastq exist many problems in practical applications, besides, fastp is an amazing utility and it has came into use in my pipelines.

I guess it is a great work that we can develop an alternative tool in the very beginning of the analysis workflow.

Once it works, maybe a new workflow of raw data processing standard is born.

Att.

The chunk should have a individual flag to identify whether encode quality

Hi, thanks for the tool! The compression ratio is IMPRESSIVE, down to 3.1%! But I got the following error using decompression...
"
~/bin/repaq -d -i haha.rfq -o ha_1.fq -O ha_2.fq
terminate called after throwing an instance of 'std::out_of_range'
what(): basic_string::substr: __pos (which is 404978952) > this->size() (which is 404978876)
Aborted (core dumped)
"