Guessing The Quality Scale In Fastq Files

Question

[Manuel]

Is there an easy way to guess the scale, given a sufficiently large FASTQ file?

The best would be some working code that I could learn from. However, both BioPerl and BioPython appear not to contain guessing code.

Answer 1

[brentp]

You read the biopython code here? That's the best explanation of the quality scores I've seen.

There's also a nice text-graphic about 2/3rd's of the way down the wikipedia page

Finally, FastQC guesses the encoding of your quality scores, so you could look at the java code.

[Manuel]

Thanks, BioPython does not hav guessing code, though, right? FastQC just looks at the lowest seen quality. I guess that's most promising, then, maybe augmented by checking an upper limit, too.

Answer 2

[Mikael Huss]

Here is a Perl script for guessing the quality scale

https://www.uppnex.uu.se/content/check-fastq-quality-score-format

[Juke34]

Here is the new link for this Perl tool: http://www.uppmax.uu.se/userscript/check-fastq-quality-score-format

It has been improved recently.

-- update --

You can find it in this repository, under this name fastq_guessMyFormat.pl: https://github.com/NBISweden/GAAS/tree/master/annotation/Tools/Util

Here is a link to download it directly.

[Yahan]

link is meanwhile broken also.

[Juke34]

Thanks,

Updated now

Answer 3

[Pierre Lindenbaum]

Does the FAST-X toolkit answer your needs ? http://hannonlab.cshl.edu/fastx_toolkit/commandline.html#fastq_quality_boxplot_usage

[Manuel]

Hm, I would like to do this programatically. I think something like the FastQC guesser looks more promising. Thanks, though.

Answer 4

[Ryan Thompson]

I wrote a Python-based FASTQ quality guesser: https://github.com/DarwinAwardWinner/fastqident It uses BioPython's FASTQ parser, so it will work on anything that is parsable by BioPython.

[Jeremy Leipzig]

i am getting 404'd

[xapple]

Looks good, but it doesn't install correctly. The module "placsupport" cannot be found in PyPI.

[Keith Callenberg]

The placsupport module can be found at https://github.com/DarwinAwardWinner/placsupport

Answer 5

[Marvin]

Isn't that solving the wrong problem? The guessing code in FastQC looks fragile, it simply looks at the smallest code used for qualities, so it depends on actually seeing low quality bases.

I believe you should get the correct encoding from extra knowledge (i.e. knowing which version of which program generated the file, say from some log file), and then convert to a well specified format (e.g. BAM) once. Please don't perpetuate the practive of guessing at the details underspecified formats.

Answer 6

[Sequencegeek]

In addition to Ryan, I have a python based fastq quality guesser as well if you would like to use it. It is just standard python (no biopython). PM if interested.