When to merge multiple fastq files into one for RNAseq analysis?

Question

[lunarskye222]

I'm a little confused on the files I am trying to analyze and so I just wanted to doublecheck on here before proceeding if that's okay:

For each RNA-seq sample (paired end sequencing), the sequencing core gave back 4 fastq files: samplenameL001_R1, samplenameL001_R2, samplenameL002_R1, samplenameL002_R2. Are the L001 and L002 separate replicates, or are they separate pieces of the same sample?

If the latter, I was thinking of combining the two fastq files (so combining the R1s together and the R2s together) before trimming them for adapters/quality. Would this be the correct thing to do? Or should I trim them all separately and combine them like so before using bowtie2 for aligning?

Or am I going about this totally wrong and if so, should I just stick to aligning them separately?

Thank you!

Answer 1

[GenoMax]

As long as the sample name matches across the lanes they are pieces of the same sample.

I was thinking of combining the two fastq files (so combining the R1s together and the R2s together) before trimming them for adapters/quality. Would this be the correct thing to do? Or should I trim them all separately and combine them like so before using bowtie2 for aligning?

You could do either but be sure to keep R1/R2 files in sync by processing them together when trimming. Once that is done, you can cat them together in the same order as described here: Concatenating fastq.gz files across lanes

You can even merge the BAM files after alignment. This will allow you to parallelize the process by processing the four sets of files in parallel.