Phred Quality Score Converter
Convert a Phred quality score into its base-calling error probability and accuracy, and see the Phred+33 and Phred+64 ASCII characters used in FASTQ files. You can also convert an error probability back to a Q score.
- Error probability
- 1.00e-2
- Phred+33 char
- 5
- Phred+64 char
- T
Enter an error probability between 0 and 1 to get the Q score.
How it works
A Phred quality score Q encodes the probability P that a base call is wrong, on a logarithmic scale: P = 10^(−Q/10), so Q = −10 × log10(P). A higher score means a more confident call — Q20 is a 1-in-100 error rate (99% accuracy), Q30 is 1-in-1000 (99.9%), and Q40 is 1-in-10000 (99.99%).
In FASTQ files the score is stored as a single ASCII character. Sanger and modern Illumina use Phred+33 (add 33 to Q and take that character), while older Illumina pipelines used Phred+64. Enter a Q score to see its probability, accuracy and both characters, or enter an error probability to get the equivalent Q.
Examples
- Q20 → error probability 1e-2 → 99% accuracy → Phred+33 char '5'.
- Q30 → 1e-3 → 99.9% accuracy → Phred+33 char '?'.
- Q40 → 1e-4 → 99.99% accuracy → Phred+33 char 'I'.
Frequently asked questions
- What is a Phred quality score?
- It is a measure of the confidence of a base call in DNA sequencing, defined as Q = −10 × log10(P) where P is the probability the call is wrong.
- What is the difference between Phred+33 and Phred+64?
- They are two ASCII encodings of the score in FASTQ files. Phred+33 (Sanger, Illumina 1.8+) adds 33 to Q; Phred+64 (older Illumina) adds 64. The correct offset is needed to decode qualities.
- What accuracy does Q30 mean?
- Q30 corresponds to an error probability of 0.001, or 99.9% base-call accuracy — one wrong base in a thousand. It is a common quality threshold for sequencing reads.
- How do I convert probability to Q?
- Use Q = −10 × log10(P). Enter the error probability in the second field and the calculator returns the Q score.