Sequence identifiers considerations when merging FASTQ files before alignment

Question

[alpanagi]

I have concatenated two FASTQ from the same sample but different lanes. I tried to pass them to basespace from illumina to do the STAR alignment but it failed saying that the samples are from different lanes (based on the sequence identifier). I updated the files to have the same lanes but still get a random error when trying some files.

Now I am setting up STAR aligner on my own PC but I was wondering what the role of the sequence identifier is when doing STAR and how should I merge these two files correctly to not affect downstream analysis.

[GenoMax]

Perhaps you should have done this in BaseSpace rather than concatenating the files.

[alpanagi]

I had no idea this could be done. This helped a lot. Thanks!

[WouterDeCoster]

but still get a random error when trying some files.

You'll have to be more specific.

[alpanagi]

I would have also liked to know but the analysis doesn't actually describe the error and support couldn't elaborate further at least until now

Answer 1

[Brian Bushnell]

The lane has nothing to do with alignment. STAR should run fine on the concatenated files when you run it yourself, and the sequence identifier will not affect downstream analysis (other than for ensuring pairing is correct or optical duplicate removal).

[GenoMax]

This has more to do with how BaseSpace implements apps (of which STAR is one I suppose).

[alpanagi]

That is ultimately what I wanted to know. Thanks!