when to merge lane replicates?

Question

[marongiu.luigi]

Dear all,

I have a set of samples each having multiple lane's replicates. When shall I merge the different replicate? BEFORE (that is, merge the fastq files) or AFTER (that is, merge the BAM files) the alignment?

Does the step of merging varies if the study is WGS or RNAseq?

I haven't found a straight answer on Google, so I am asking.

Thank you

[Devon Ryan]

Is this just the same library being loaded on the multiple lanes or are these actually different libraries?

Answer 1

[GenoMax]

If the same sample ran on multiple lanes as a part of pool you can merge the data at fastq stage (people will tell you to add RG etc., it is your call). Sequencing facilities should use --no-lane-splitting option when using bcl2fastq to avoid this issue.

[marongiu.luigi]

perfect, thank you! now it is clear