How can I create a ciclo for in bash to merge each R1 fastq and R2 fastq in a single-end file?

Question

[lorenzo.paci]

Good day everyone,

I have this file set:

1_S1_R1.fastq.gz
1_S1_R2.fastq.gz
2_S2_R1.fastq.gz
2_S2_R2.fastq.gz
3_S3_R1.fastq.gz
3_S3_R2.fastq.gz

I would want create a ciclo for with the software SeqPrep in which each couple of reads file will merge together; for example 1_S1_R1.fastq.gz + 1_S1_R2.fastq.gz = 1_S1_merged.fastq.gz and so on for all the files in the directory. Is this possible? Or am I just imagine?

Best regards
Lorenzo

[Marco Pannone]

Pure fantasy of a bioinformatic world that does not exist. You want to basically merge paired-end data into single-end, this is probably a crime in bioinformatics.

[ATpoint]

@OP, do you mean an interleaved file, so for each read you have alternative R1/R2/R1/R2... in the same file, or do you simply want to take entire R1 file and concatenate R2 to it (the latter would be, as said, questionable)?

Answer 1

[GenoMax]

ciclo = loop?

Merging can mean different things. What you are asking for 1_S1_R1.fastq.gz + 1_S1_R2.fastq.gz = 1_S1_merged.fastq.gz is concatenation of the files end to end. Very few packages can use data in this format.

More common usage would be to either interleave R1/R2 data files (described here Combine paired-end fastq files ) or to actually merge individual reads to create a longer single read representation (if the reads overlap in the middle, when number of cycles of sequencing is > (insert size/2)) using a program like bbmerge.sh or FLASH Insert size provided by bbmerge