Question: deML 'rgqual' parameter
gravatar for ovon
16 months ago by
ovon20 wrote:

Could someone familiar with deML explain the 'rgqual' parameter a bit better? The default value listed for rgqual in the documentation is 80, but with the quality scoring system in all current Illumina data (phred 33), quality scores this high don't exist. Is this a different type of quality score? If so, could you elaborate a little on how it relates to PHRED quality, how it is calculated, and why the default is 80?

ADD COMMENTlink modified 16 months ago • written 16 months ago by ovon20
gravatar for Gabriel R.
16 months ago by
Gabriel R.2.7k
Danmarks Tekniske Universitet
Gabriel R.2.7k wrote:

The rgqual refers to the Z_0 score. Z_0 is the likelihood for the top (most likely) read group on a PHRED scale. Considering the sequence of this read group as the template to sequence and using the principle of independence between bases, this likelihood is computed as:

Z_0= -10 * log_10{ PROD_{1...length indices} P(base read i|base template i)  }

The higher the probability of assignment, the lower this score becomes. If say the probability is almost 1 (perfect), the score Z_0 will be around 0. Conversely, if the probability of assignment is dismal around 0, the score will be 200 or 300. To compute this, we use quality scores but the correspondence is not 1 to 1, it depends on the template and sequenced bases.

I think the confusion stems from the use of the PHRED scale. PHRED is a mathematical formula and does not always refer to quality scores. Mapping quality is computed on a PHRED scale for instance.

ADD COMMENTlink written 16 months ago by Gabriel R.2.7k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1548 users visited in the last hour