ReadDepth

Read depth, or sequencing depth, is a measure of how many sequencing reads cover a given position in a reference genome. Per-base read depth equals the number of reads that align to that base. The average depth across the genome is the total number of bases sequenced that align to the reference divided by the genome length; in targeted sequencing, depth is reported as mean depth across targeted regions.

Read depth is typically expressed in times, denoted as x. For example, 30x depth means each base is covered, on average, by 30 reads. Because coverage varies, the distribution is often modeled as Poisson or negative binomial, with biases due to GC content, duplicates, and mappability.

Factors affecting depth include sequencing technology, library preparation, read length, and target design. High depth increases confidence in variant calls, especially for detecting rare variants or mosaicism, but with diminishing returns and higher cost. Depth is related to, but distinct from, coverage, which refers to the proportion of the genome covered at or above a given depth threshold (e.g., 20x).

Applications include variant discovery in whole-genome and exome sequencing, copy-number variation inference from read depth, and quality control in sequencing projects. Typical WGS depth ranges ~30–40x; exome sequencing often aims for 80–200x in targeted regions, depending on goals.

Limitations include uneven depth due to sequence context, regions where reads map poorly, and artifacts from duplicates or pseudogenes. Reporting often includes mean, median, and the proportion of bases at or above certain depths, to convey depth distribution.

Accurate assessment of read depth is essential for interpreting sequencing data and for planning sequencing strategies.