I have two fastq files. I want to use bwa to produce two sam files as this is more efficent for me than merging the fastq files and running bwa on the bigger file. If I run bwa separately on the fastq files and merge the sam files later, will the result be different? (I don't know if this is an important detail, but I know a priori that the fastq files have reads from separate chromosomes.)
Individual reads are mapped independently of each other. So mapping the reads in two chunks should produce the same mappings as a single file. However, I wouldn't expect mapping runs to always be identical as there is a random element to mapping repetitive hits.