Sequencing reads and how a genome is assembled

A read is a reported DNA sequence from a sampled fragment. Reads may be short or long, depending on the method. They are not automatically a finished genome. Assembly uses shared sequence between reads to reconstruct longer regions; some assemblies remain incomplete.

Coverage depth counts the reads covering a base. Mean depth averages this across the target. Tenfold mean coverage does not mean ten reads at every base. Extra independent reads can improve consensus accuracy when errors are random. They need not remove systematic errors or fill every gap.

A read wholly inside an identical repeat may fit several copies. More internal reads do not identify its location. A read spanning the repeat and reaching distinctive sequence on both sides can provide a placement link. Other linking information can also help.

Choose reads for the task and available methods. If two options meet the accuracy and assembly needs, compare their stated costs at suitable coverage. For repeats, check whether reads reach distinctive flanks. Read length alone does not determine cost or accuracy.

Unique sequence reduces placement ambiguity but does not guarantee a complete assembly. Enough usable reads and connections are still needed. Separate base errors, uncertain placement and missing sequence when deciding what data to collect.