Mismatch repair corrects errors that escape proofreading during DNA replication. These errors typically involve the insertion of the wrong base—such as a guanine paired with a thymine instead of a cytosine.

The mismatch repair system identifies the incorrect base by detecting distortions in the DNA helix. Enzymes remove a section of the newly synthesized strand that contains the error. DNA polymerase then fills in the correct sequence, and DNA ligase seals the strand.

Nucleotide excision repair handles larger, bulkier lesions that distort the shape of the DNA helix. A common example is a thymine dimer, which forms when UV radiation causes two adjacent thymine bases to bond together.

In this process, a section of the DNA strand surrounding the damaged area is removed by a group of enzymes. The gap is then filled in with the correct nucleotides using the undamaged strand as a template. 

Sometimes, during replication, the DNA polymerase encounters a lesion that it cannot copy across. Rather than stall completely, the cell uses a strategy called DNA damage bypass to continue replication past the damaged site.

In this process, the replication machinery skips over the lesion and restarts downstream, leaving behind a single-stranded gap opposite the damaged region. Later, the cell uses strand exchange to repair this gap. A portion of the undamaged parental strand is used as a template to fill in the missing section. Once the correct sequence is restored, the replication machinery can rejoin the segment, and the gap is sealed by DNA ligase.

Double-strand breaks are one of the most dangerous types of DNA damage. If both strands of the helix are broken, the genetic information in that region can be lost unless it is accurately restored. Double-strand gap repair is a form of homologous recombination used to repair such breaks, especially when a sister chromatid is available as a template.

The repair begins with strand invasion. One of the broken DNA ends invades the complementary region on the sister chromatid. This displaces the original strand on the template and forms a structure called a D-loop. DNA polymerase then uses the invaded strand as a template to extend and elongate the broken strand, copying the missing information.

Once elongation has proceeded far enough to restore the missing region, the invading strand is released from the template and reanneals with its original partner strand. DNA polymerase fills in any remaining gaps, and ligase seals the backbone.