About 50% of the human genome is composed of repetitive DNA elements. These are sequences that occur many times, either as repeated blocks, mobile genetic elements, or other duplicated patterns distributed throughout the genome.
Much of this repetitive DNA comes from transposable elements, sometimes called “jumping genes.” Important categories include Alu elements, LINEs, SINEs, and endogenous retrovirus-related sequences. Many no longer move through the genome, but they remain as molecular records of past evolutionary events.
Repetitive DNA was a major technical obstacle during the Human Genome Project. Sequencing machines read relatively short fragments, and identical or near-identical repeats made it difficult for assembly software to determine each fragment's precise location. This is one reason the original project focused primarily on euchromatic DNA and left some repetitive regions unfinished.
The estimate is approximate because the percentage depends on how researchers define and classify repetitive elements. It should not be confused with the percentage of DNA that directly codes for proteins: protein-coding sequences make up only a tiny fraction of the genome, while much of the rest includes repetitive and other noncoding DNA.