HISAT2 output error

Hi,

I have recently been processing a large number of fastq files for RNA-seq analysis (150bp, paired-end, Illumina platform) following the tutorial below:

(Hands-on: 1: RNA-Seq reads to counts / 1: RNA-Seq reads to counts / Transcriptomics)

I am currently at the HISAT2 for mapping the reads to the hg38 genome but have encountered this error message for several runs:

Error, fewer reads in file specified with -1 than in file specified with -2 terminate called after throwing an instance of ‘int’ (ERR): hisat2-align died with signal 6 (ABRT) (core dumped)…

Some files are able to run successfully while others are not. I have checked the files via FastQC and MultiQC and there are the same number of reads in the paired files. The quality and QC reports of all the files look roughly similar so I’m not sure why some files are able to run successfully while others are not. The files are also all in fastqsanger.gz format as cutadapt outputs.

Would appreciate any advice on what else I can do or check so that I don’t have to wait hours for a failed run.

Thanks!

Hi @wcm
The error message suggests that non-paired reads were used. It can be accidental submission of files from different samples or reads were trimmed as SE, not PE, or something else.
If you trimmed the reads in PE mode, check job setup of the failed job(s) using info (i) or Re-run icons. If the files are from the same sample, you can share the history and post the link here, so someone can check what is going on.

History sharing: History options (three horizontal bars icon at the top right corner of the history panel) > Share or Publish > Make the history accessible (in the middle window) > Copy and paste the URL in reply.

Kind regards,
Igor

1 Like

Hi Igor,

Thanks for the reply. I did indeed trim the samples as SE thinking it would be easier to process them as a collection as they share the same adapter sequences. Did not realize that this would affect the output. I’ll try rerunning the samples trimming the reads in PE mode.

Hi @wcm
I am glad you figured it out.
Files with properly paired reads F and R should have identical number of sequences sorted in the same order. Trimming in PE mode usually produces several files: files with properly paired reads and files with “orphan” reads, when a pair is missing.
Kind regards,
Igor