Does this account for paired-end reads?

Count each mate as one read. Two 150 bp mates of a pair are two reads of 150 bp, i.e. 300 sequenced bases.

Why is my real coverage lower than this estimate?

This is a theoretical mean. Duplicate reads, off-target reads, low-quality trimming and uneven coverage all reduce usable depth.

Sequencing coverage calculator

Work out mean depth of coverage from genome size, read length and read count — or the number of reads you need to hit a target depth.

How it works

Formula

depth = (read length × read count) ÷ genome size

Worked example

A human genome is about 3,000,000,000 bp. With 150 bp reads and 600,000,000 reads: (150 × 600,000,000) ÷ 3,000,000,000 = 30× mean depth.

When to use it

Before a run, to choose how many reads to order for the depth your application needs (e.g. ~30× for human germline variant calling, 100×+ for somatic or low-frequency variants). After a run, to sanity-check the depth you actually got.

Sensible defaults

Defaults model a 30× human whole-genome sample on 2 × 150 bp chemistry. Swap in your own genome size for bacteria, exomes or panels.

FAQ

Does this account for paired-end reads?: Count each mate as one read. Two 150 bp mates of a pair are two reads of 150 bp, i.e. 300 sequenced bases.
Why is my real coverage lower than this estimate?: This is a theoretical mean. Duplicate reads, off-target reads, low-quality trimming and uneven coverage all reduce usable depth.