Does bowtie2 still require fastqsanger format?

Question

[bxia]

Hi,

I just found a old tutorial said galaxy bowtie2 require fastqsanger format,

Does it still a requirement for bowtie2-2.2.9 just downloaded from bowtie2 website?

Thanks

[GenoMax]

If you are using data produced in the last 3-4 years it will already be in sanger fastq format. Are you using galaxy or stand-alone bowtie2?

[bxia]

I am using stand alone bowtie2

[GenoMax]

Then nothing needs to be done. Go ahead scan/trim and align.

[bxia]

Do you know how long it will take? My laptop is i7-3610QM with 8Gb RAM, have waited for 2 hrs for the first 3 million reads, is it normal?

[GenoMax]

There is a lot of I/O involved with alignments since you are reading/writing large files. On a laptop (even if you have an SSD) it is going to take time to go through a large dataset. With bowtie2 8GB of RAM should be adequate but with some other programs (e.g. STAR which needs significantly more RAM) you may not be able to use your laptop.

Answer 1

[Steven Lakin]

You can specify what type of quality scores were used to generate your fastq files with the options; it's in the manual. The majority of fastq files should have Phred33 offset now.

--phred33
Input qualities are ASCII chars equal to the Phred quality plus 33. This is also called the "Phred+33" encoding, which is used by the very latest Illumina pipelines.

--phred64
Input qualities are ASCII chars equal to the Phred quality plus 64. This is also called the "Phred+64" encoding.

--solexa-quals
Convert input qualities from Solexa (which can be negative) to Phred (which can't). This scheme was used in older Illumina GA Pipeline versions (prior to 1.3). Default: off.

--int-quals
Quality values are represented in the read input file as space-separated ASCII integers, e.g., 40 40 30 40..., rather than ASCII characters, e.g., II?I.... Integers are treated as being on the Phred quality scale unless --solexa-quals is also specified. Default: off.