How to align non overlapping fastq files

Question

[Rongen]

Hi, We have sequenced 250 bp read length library at 100 X 2 cycles on Miseq by mistake, because of which the R1 and R2 fastq files are of 100 bp in length each and are non-overlapping. How to align these fastq files to reference genome. Can we align R1 as single end sequencing file to generate .BAM for R1. Also reverse complement R2 and do the same. Later we can merge two .BAM files for downstream processing. Any suggestions?? Thanks in advance.

Answer 1

[WouterDeCoster]

There is absolutely no need to have overlapping reads for alignment. Most (if not all) aligners take read pairs (two files) as input.

[Brian Bushnell]

To expand on that, you generally need to tell the aligner the locations of the read files and reference (often you have to index the reference in one step, and then align the two read files together in a second step), and it will take care of everything else. Do not align them independently, and do not reverse-complement read 2, and definitely do not concatenate the files or fuse the reads together (without a read-merging program designed for that purpose); any of those will confuse the aligner and yield inferior output.