Mapping multiple FASTQ data types using BWA?

[Genomics101]

Greetings!

I plan to do mapping using multiple types of Illumina FASTQ files for the same sample.

I have one set of FASTQs made from a library with 400bp inserts, 100bp reads (trimmed for quality), and processed with the 1.5 pipeline; the second set is 600bp inserts, 250bp reads (trimmed for quality), and processed with the 1.9 pipeline. Because of the trimming process, I have paired-end and single-end FASTQs for each sample as well.

As a result I have 8 FASTQs for mapping, e.g. -

Insert size 400:
Sample1_Illumina_1.5_100bp_R1.fastq.pe
Sample1_Illumina_1.5_100bp_R2.fastq.pe
Sample1_Illumina_1.5_100bp_R1.fastq.se
Sample1_Illumina_1.5_100bp_R2.fastq.se

Insert size 600:
Sample1_Illumina_1.9_250bp_R1.fastq.pe
Sample1_Illumina_1.9_250bp_R2.fastq.pe
Sample1_Illumina_1.9_250bp_R1.fastq.se
Sample1_Illumina_1.9_250bp_R2.fastq.se

I have only been mapping with the 100bp/1.5 so far, using BWA, but reading it looks like I should use BWA-MEM if I am using the longer reads. Is it possible to map all of these together?

Thanks!

[adamdeluca]

I would map separately for each insert size, then merge the resulting bams for further analysis.

[dpryan]

No, don't map them together. Also, bwa mem doesn't support illumina 1.5.