Illumina Q score

Question

[A_heath]

Hi all,

I have Illumina sequencing results of a bacterial genome and a quality score of 35.89 is associated with these data. I know that a quality score of 30 is 99.99% of base calling accuracy based on this but what about the meaning of 35.89? Is there a scale used for quality score?

Thank you very much for your much appreciated help on this!

[Friederike]

Can you elaborate a bit on how you obtained that quality score? Is that the average quality across all Illumina reads? Was it determined by a specific tool? If yes, how?

[A_heath]

Thank you for your help. This quality score that I mentionned is the phred quality score. I would like to interpret this data properly.

Answer 1

[Friederike]

The phred score describes the probability of a base call being incorrect. Wikipedia has a great explanation including the formula, which is -10xlog10(probability of being wrong).

phred = 35.89 # = -10*log10(probability)
35.89/-10 = -3.589
probability = 10^-3.589 = 0.0002576321

Therefore, a base call with a Phred score of 35.89 has a 0.02% chance of being wrong. Or, in other words, 0.02% of bases with a Phred score of 35.89 will have been wrongly assigned.

[A_heath]

Thank you very much for this clear explanation!!