Hi, I am experiencing a reproducible hang with fastp during automatic adapter detection on paired-end Illumina shotgun metagenomic data.
The process stops after:
Detecting adapter sequence for read1...
No adapter detected for read1
Detecting adapter sequence for read2...
No adapter detected for read2
After this point, fastp remains running indefinitely and the output FASTQ files remain at 0 bytes.
Environment
- fastp version: 1.3.7
- OS: Linux (CentOS/cluster environment)
- Architecture: x86_64
- Input: paired-end Illumina shotgun metagenomic data
- Read length: 150 bp
- Compression: gzip
- Compute environment: SLURM cluster
- CPUs allocated: 35
- RAM allocated: 500 GB
The issue occurs with the following command:
~/bin/fastp \
-i "$R1" \
-I "$R2" \
-o "${sample}_1.clean.fq.gz" \
-O "${sample}_2.clean.fq.gz" \
--detect_adapter_for_pe \
--cut_front \
--cut_tail \
--cut_mean_quality 20 \
--length_required 50 \
--thread 35 \
--html "${sample}.fastp.html" \
--json "${sample}.fastp.json"
The process stops at the adapter detection stage:
Detecting adapter sequence for read1...
No adapter detected for read1
Detecting adapter sequence for read2...
No adapter detected for read2
The same data can be processed successfully when automatic adapter detection is disabled.
For example, using a subset of 1,000,000 paired reads:
fastp \
-i /tmp/EC18a_test_R1.fq.gz \
-I /tmp/EC18a_test_R2.fq.gz \
-o /tmp/test_R1.clean.fq.gz \
-O /tmp/test_R2.clean.fq.gz \
--thread 2 \
--disable_adapter_trimming
This completed successfully in approximately 20 seconds.
Could this be related to the automatic adapter detection algorithm for paired-end reads?
Thank you in advance for your help,
Best regards,
Guillaume
Hi, I am experiencing a reproducible hang with fastp during automatic adapter detection on paired-end Illumina shotgun metagenomic data.
The process stops after:
Detecting adapter sequence for read1...
No adapter detected for read1
Detecting adapter sequence for read2...
No adapter detected for read2
After this point, fastp remains running indefinitely and the output FASTQ files remain at 0 bytes.
Environment
The issue occurs with the following command:
The process stops at the adapter detection stage:
The same data can be processed successfully when automatic adapter detection is disabled.
For example, using a subset of 1,000,000 paired reads:
fastp \ -i /tmp/EC18a_test_R1.fq.gz \ -I /tmp/EC18a_test_R2.fq.gz \ -o /tmp/test_R1.clean.fq.gz \ -O /tmp/test_R2.clean.fq.gz \ --thread 2 \ --disable_adapter_trimmingThis completed successfully in approximately 20 seconds.
Could this be related to the automatic adapter detection algorithm for paired-end reads?
Thank you in advance for your help,
Best regards,
Guillaume