Hello,
I’m new to using the HUMAnN and have a question about the best practice for preparing input files.
I have a set of metagenomic samples, and for each patient, the raw data consists of multiple paired-end (R1/R2) FASTQ files as well as some single-end FASTQ files. Based on my previous experience, my initial approach was to first use seqtk mergepe to interleave all the paired-end reads. Then, I used the cat command to concatenate the resulting interleaved file with all of the single-end read files into a single input file for MetaPhlAn and HUMAnN.
However, I saw a discussion here today suggesting that one can simply use cat to concatenate all the raw FASTQ files (both paired-end and single-end) directly into one large file.
I’m very curious if there is a significant difference in accuracy between these two methods. Is one approach better than the other, or are both acceptable for use with HUMAnN?
Any help or clarification on this would be greatly appreciated!
Thank you!