quality filtering of Miseq dataset

Question

[User 4014]

Hi folks,

I am new to NGS, but I am about to work on amplicon sequencing with Miseq (PE). I am now studying an article from Balint et al. (2014) An Illumina metabarcoding pepline for fungi, which mentions 'It is important to preserve the order of the reads in both forward and reverse read files: the paired-end read assembler needs corresponding read orders in both files'.

This might be an innocent question, but I would like to ask how I obtain the files with the right order of the reads preserved in both forward and reverse read files? Do I get the files from a sequencing facility with specific requests or do I need a program or a script to work on afterwards?

Thank you in advance!

Johan

[User 4014]

Thank you very much for super quick replies! :)

Answer 1

[Dan D]

Your FASTQ data will come off the MiSeq like this:

[Sample Name]_S[Sample number]_L001_R[read number]_001.fastq.gz
#                                   ^^^^^^^^^^^^^^

Read 1 data will have R1 for the highlighted portion of the file name.

Read 2 data will have R2 for the highlighted portion.

What the paper is saying is not to combine those files before feeding them to the assembler. There are some (mostly old) programs (trimmers, groomers, etc) that combine paired-end data into a single file (this is sometimes termed an interleaved dataset) in the course of processing.

Answer 2

[Brian Bushnell]

If you have paired reads, they should come from the facility in two files, both in the same order (or in one file with read 1 and read 2 alternating). To keep them in the same order during filtering or trimming, you just need to use a pair-aware program such as BBDuk that processes both reads at the same time.