Unable to read file magic number

[GaojieLi-dot]

When I use bismark to map DNA methylation data, the log file shows this error. At the same time, the mapping is also very low.
error infotmation:
Unable to read file magic number
is_zstd_file: unable to read magic number
Unable to read file magic number
is_zstd_file: unable to read magic number
Unable to read file magic number
is_zstd_file: unable to read magic number

report:
Sequences analysed in total: 2401399
Number of alignments with a unique best hit from the different alignments: 772217
Mapping efficiency: 32.2%
Sequences with no alignments under any condition: 1435910
Sequences did not map uniquely: 193272
Sequences which were discarded because genomic sequence could not be extracted: 0

Number of sequences with unique best (first) alignment came from the bowtie output:
CT/CT: 176210 ((converted) top strand)
CT/GA: 175036 ((converted) bottom strand)
GA/CT: 210189 (complementary to (converted) top strand)
GA/GA: 210782 (complementary to (converted) bottom strand)

command:
bismark -q -N 1 --parallel 5 --non_directional --bam --temp_dir ./ -un ./ --bowtie2 --genome hg38 fastq.gz -o ./

[FelixKrueger]

Googling this error isn't overly helpful, some suggest that it could have to do with malformatteted input files?

Regarding the low mapping, this can have a number of reasons, most notably appropriate trimming. Which type of sequencing is is, what did you do regarding trimming. And: do you have enough system resources to support --parallel 5? (e.g. >60GB of RAM?)

[ahanden]

That error comes from bowtie2. It usually shows up if your input fastq files end in .gz but aren't actually gzipped. Double check the contents of your fastq.gz file.