BWA MEM alignment

Question

[Qboy]

Happy New Year,

I have a question regarding BWA (this is my first time doing it). I have a nice working script for paired-end reads. However, when I converted my SRA files to Fastq, I used fastq-dump --split-3, because I am also interested in small reads due to the quality of DNA I am working on now.

Questions:

1. Fastq generated three files fastq, fastq_1, fastq_2. Is it possible to put all three to bwa mem? Or what is the best strategy in this case?
2. I have some files which are single reads. Can I use BWA MEM for single reads too (quality of DNA is low meaning I have more short reads than long, and the longest is about 100 bp)?

Answer 1

[Istvan Albert]

The third file usually contains the index sequences associated with sample multiplexing. There is no need to align those.

You can align both single-end and paired-end reads, just not in the same invocation of the command. Basically you would need to run it separately for each data.

[Qboy]

Thank you so much! I think I am now a bit more confident with my next step! Merci

Answer 2

[Raman2]

1. Use fastq_1 and fastq_2 for bwa-mem. (btw, fastq-dump is an older tool. The newer fasterq-dump is faster, multi-threaded and can download and generate fastq files from SRA from a single command line)
2. Yes, bwa-mem works for single end data