thebench.bio #1 Ecosystem for the Bio-Tech Industry
All NGS questions

Question 11 of 20 Fundamentals Interpretation

What does a Phred quality score of Q30 mean?

Short answer

Q30 means the sequencer estimates a 1 in 1,000 chance that the base call is wrong, or 99.9 percent accuracy. The scale is logarithmic, Q = -10 log10 P, so every 10 points is a tenfold change in error probability.

Why

The Phred score turns an error probability into a small whole number. Q10 means 1 in 10 bases wrong, Q20 means 1 in 100, Q30 1 in 1,000 and Q40 1 in 10,000. Illumina's technical note explains that the scores come from lookup tables calibrated on large datasets of known sequence, and shows predicted and observed quality agreeing closely.

It also notes why Q30 became the benchmark. At Q20, a 100 base read will likely contain one error, while at Q30 nearly all reads are error-free. Low scores raise false-positive variant calls, which is why quality is checked before calling.

In practice, a run summary reports the percent of bases at or above Q30, and FASTQ files store each base's score as one ASCII character. A score is the instrument's estimate, not a measured truth. It does not capture errors made before sequencing, such as PCR mistakes or DNA damage during library preparation, which appear with high quality scores.

It also says nothing about whether the read maps to the right place, which is a separate mapping quality.

A strong answer covers

  • Q = -10 log10 of the error probability
  • Q30 is 1 error in 1,000, or 99.9 percent accuracy
  • Each 10 points is a tenfold change
  • Scores are calibrated estimates from the instrument
  • Library errors such as PCR mistakes are not reflected

Common mistakes

  • Reading Q30 as 30 percent error or 30 errors per read
  • Assuming high base quality rules out PCR or damage artifacts
  • Confusing base quality with mapping quality